AU737569B2 - Anti-alphav-beta3 recombinant human antibodies, nucleic acids encoding same and methods of use - Google Patents

Description

1 ANTI-a, RECOMBINANT HUMAN ANTIBODIES, NUCLEIC ACIDS ENCODING SAME AND METHODS OF USE BACKGROUND OF THE INVENTION The present invention relates generally to integrin mediated diseases and, more particularly, to nucleic acids encoding av 3-inhibitory monoclonal antibodies and to CDR grafted af3,-inhibitory antibodies for the therapeutic treatment of av 3 -mediated diseases.

Integrins are a class of cell adhesion receptors that mediate both cell-cell and cellextracellular matrix adhesion events. Integrins consist :of heterodimeric polypeptides where a single a chain polypeptide noncovalently associates with a single P chain. There are now about 14 distinct a chain polypeptides and at least about 8 different P chain polypeptides which constitute the integrin family of cell adhesion receptors. In general, different binding specificities and tissue distributions are derived from unique combinations of the a and 3 chain polypeptides or integrin subunits. The family to which a particular integrin is associated with is usually characterized by the 3 subunit. However, the ligand binding activity of the integrin is largely influenced by the a subunit. For example, vitronectin binding integrins contain the a, integrin subunit.

\\BRIS\home$\IsabelH\Speci\34980-doc WO 98/33919 PCT/US98/01826 2 It is now known that the vitronectin binding integrins consist of at least three different c, containing integrins. These a, containing integrins include av3, avP and all of which exhibit different ligand binding specificities. For example, in addition to vitronectin, 0cP 3 binds to a large variety of extracellular matrix proteins including fibronectin, fibrinogen, laminin, thrombospondin, von Willebrand factor, collagen, osteopontin and bone sialoprotein I.

The integrin aCj3 binds to fibronectin, osteopontin and vitronectin whereas aL is known to bind to vitronectin and osteopontin.

As cell adhesion receptors, integrins are involved in a variety of physiological processes including, for example, cell attachment, cell migration and cell proliferation. Different integrins play different roles in each of these biological processes and the inappropriate regulation of their function or activity can lead to various pathological conditions.

For example, inappropriate endothelial cell proliferation during neovascularization of a tumor has been found to be mediated by cells expressing vitronectin binding integrins. In this regard, the inhibition of the vitronectin-binding integrin a,3 also inhibits this process of tumor neovascularization. By this same criteria, 043 has also been shown to mediate the abnormal cell proliferation associated with restenosis and granulation tissue development in cutaneous wounds, for example. Additional diseases or pathological states mediated or influenced by a v3 include, for example, metastasis, osteoporosis, age-related macular degeneration and diabetic retinopathy, and inflammatory diseases such as rheumatoid arthritis and psoriasis.

Thus, agents which can specifically inhibit vitronectin- WO 98/33919 PCTIUS98/01826 3 binding integrins would be valuable for the therapeutic treatment of diseases.

Many integrins mediate their cell adhesive functions by recognizing the tripeptide sequence Arg-Gly- Asp (RGD) found within a large number of extracellular matrix proteins. A variety of approaches have attempted to model agents after this sequence to target a particular integrin-mediated pathology. Such approaches include, for example, the use of RGD-containing peptides and peptide analogues which rely on specificity to be conferred by the sequences flanking the RGD core tripeptide sequence. Although there has been some limited success, most RGD-based inhibitors have been shown to be, at most, selective for the targeted integrin and therefore exhibit some cross-reactivity to other non-targeted integrins. Such cross-reactive inhibitors therefore lack the specificity required for use as an efficacious therapeutic. This is particularly true for previously identified inhibitors of the integrin cUj3.

Monoclonal antibodies on the other hand exhibit the specificity required to be used as an effective therapeutic. Antibodies also have the advantage in that they can be routinely generated against essentially any desired antigen. Moreover, with the development of combinatorial libraries, antibodies can now be produced faster and more efficiently than by previously used methods within the art. The use of combinatorial methodology also allows for the selection of the desired antibody along with the simultaneous isolation of the encoding heavy and light chain nucleic acids. Thus, further modification can be performed to the combinatorial antibody without the incorporation of an additional cloning step.

WO 98/33919 PCT/US98/01826 4 Regardless of the potential advantages associated with the use of monoclonal antibodies as therapeutics, these molecules nevertheless have the drawback in that they are almost exclusively derived from non-human mammalian organisms. Therefore, their use as therapeutics is limited by the fact that they will normally elicit a host immune response. Methods for substituting the antigen binding site or complementarity determining regions (CDRs) of the non-human antibody into a human framework have been described. Such methods vary in terms of which amino acid residues should be substituted as the CDR as well as which framework residues should be changed to maintain binding specificity. In this regard, it is understood that proper orientation of the 3 sheet architecture, correct packing of the heavy and light chain interface and appropriate conformation of the CDRs are all important for preserving antigen specificity and affinity within the grafted antibody. However, all of these methods require knowledge of the nucleotide and amino acid sequence of the non-human antibody and the availability of an appropriately modeled human framework.

Thus, there exists a need for the availability of nucleic acids encoding integrin inhibitory antibodies which can be used as compatible therapeutics in humans.

For a~I3-mediated diseases, the present invention satisfies this need and provides related advantages as well.

SUMMARY OF THE INVENTION The invention provides a Vitaxin antibody and a LM609 grafted antibody exhibiting selective binding affinity to UAf 3 The Vitaxin antibody consists of at WO 98/33919 PCT/US98/01826 least one Vitaxin heavy chain polypeptide and at least one Vitaxin light chain polypeptide or functional fragments thereof. Also provided are the Vitaxin heavy and light chain polypeptides and functional fragments.

The LM609 grafted antibody consists of at least one LM609 CDR grafted heavy chain polypeptide and at least one LM609 CDR grafted light chain polypeptide or functional fragment thereof. The invention additionally provides a high affinity LM609 grafted antibody comprising one or more CDRs having at least one amino acid substitution, where the oAP3 binding activity of the high affinity LM609 grafted antibody is enhanced. Nucleic acids encoding Vitaxin and LM609 grafted heavy and light chains as well as nucleic acids encoding the parental non-human antibody LM609 are additionally provided. Functional fragments of such encoding nucleic acids are similarly provided. The invention also provides a method of inhibiting a function of avf 3 The method consists of contacting av3 with Vitaxin or a LM609 grafted antibody or functional fragments thereof under conditions which allow binding to vfP 3 Finally, the invention-provides for a method of treating an av3-mediated disease. The method consists of administering an effective amount of Vitaxin or a LM609 grafted antibody or functional fragment thereof under conditions which allow binding to oP3.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 shows the nucleotide and deduced amino acid sequence of the variable region of the antibody Vitaxin. Figure 1A shows the nucleotide and deduced amino acid sequences for the Vitaxin heavy chain variable region (Glnl-Serll7; SEQ ID NOS:1 and 2, respectively) while Figure 1B shows the nucleotide and deduced amino WO 98/33919 PCT/US98/01826 6 acid sequences for the Vitaxin light chain variable region (Glul-Lysl07; SEQ ID NOS:3 and 4, respectively).

Figure 2 shows the nucleotide and deduced amino acid sequence of the variable region of the monoclonal antibody LM609. Figure 2A shows the nucleotide and deduced amino acid sequence of the LM609 heavy chain variable region (SEQ ID NOS:5 and 6, respectively). The variable region extends from amino acid Glul to Alall7.

Figure 2B shows the nucleotide and deduced amino acid sequence of the LM609 light chain variable region (SEQ ID NOS:7 and 8, respectively). The variable region of the light chain extends from amino acid Aspl to Lysl07.

Figure 3 shows the competitive inhibition of LM609 IgG binding to the integrin avb3 with recombinant LM609 Fab. Soluble recombinant murine LM609 Fab fragments were prepared from periplasmic fractions of M13 bacteriophage clones muLM609M13 12 and muLM609M13 29.

The periplasm samples were serially diluted, mixed with either 1 ng/ml, 5 ng/ml, or 50 ng/ml of LM609 IgG and then incubated in 96 well plates coated with purified aC 3 Plates were washed and bound LM609 IgG detected with goat anti-murine Fc specific antibody conjugated to alkaline phosphatase. Fab produced by clone muLM609M13 12 inhibits both 1 ng/ml and 5 ng/ml LM609 IgG binding at all concentrations of Fab greater than 1:27 dilution.

Figure 4 shows the characterization of Vitaxin binding specificity. Figure 4A shows specific binding of Vitaxin to the integrin av3 compared to integrins IIbP3 and av, 5 Figure 4B shows the competitive inhibition of LM609 binding to PV03 by Vitaxin. Figure 4C shows the competitive inhibition of fibrinogen binding to acv 3 by Vitaxin.

WO 98/33919 PCT/US98/01826 7 Figure 5 shows the inhibition of c~3 3 -mediated cell attachment (5A) and migration (5B) by Vitaxin.

Figure 6 shows the reduction in tumor growth due to Vitaxin mediated inhibition of neovascularization.

Figure 6A shows the inhibition of the c~v3-negative Fg and HEp-3 human tumor fragments grown on chick chorioallantoic membranes (CAMs) following Vitaxin treatment. Figure 6B shows the growth inhibition of Vx2 carcinomas implanted subcutaneously in rabbits at two different Vitaxin doses administered 1 day post implantation. Figure 6C similarly shows Vx2 tumor growth inhibition as in Figure 6B, except that four different Vitaxin doses were administered beginning at 7 days post implantation.

Figure 7 shows the nucleotide and deduced amino acid sequence of the light chain variable region of the LM609 grafted antibody fragment (Glul-Lysl07; SEQ ID NOS:31 and 32, respectively). Position 49 of the light chain variable region can at least be either Arg or Met.

The nucleotide and deduced amino acid sequence of the heavy chain variable region of the LM609 grafted antibody fragment is shown in Figure 1A (SEQ ID NOS:1 and 2, respectively).

Figure 8 shows the titration of LM609 grafted antibody variants and LM609 grafted Fab on immobilized ovP3. Bacterial cell lysates containing LM609 grafted antibody (closed circles), LM609 grafted antibody variants with improved affinity isolated from the primary libraries (S102, closed squares; Y100, open squares; and Y101, open triangles) or from combinatorial libraries (closed triangles), or an irrelevant Fab (open circles) were titrated on immobilized avP3.

WO 98/33919 PCT/US98/01826 8 Figure 9 shows the construction of combinatorial libraries of enhanced LM609 grafted antibody variants containing multiple amino acid substitutions.

Figure 10 shows the inhibition of fibrinogen binding to aA3 by LM609 grafted antibody variants.

Figure 10A shows inhibition of fibrinogen binding to immobilized aP, 3 Figure 1B shows correlation of affinity of antibody variants with inhibition of fibrinogen binding.

Figure 11 shows the inhibition of M21 human melanoma cell adhesion to fibrinogen by LM609 grafted antibody variants. Cell binding to 10 pg/ml fibrinogencoated substrate was assessed in the presence of various concentrations of LM609 grafted Fab (closed triangles) or the enhanced LM609 grafted Fabs S102 (open circles), G102 (closed circles), or C37 (open triangles).

DETAILED DESCRIPTION OF THE INVENTION The invention is directed to nucleic acids encoding the monoclonal antibody (MAb) LM609. This antibody specifically recognizes the integrin 0Av3 and inhibits its functional activity. The invention is also directed to nucleic acids encoding and to polypeptides comprising non-murine grafted forms of LM609. These grafted antibodies retain the binding specificity and inhibitory activity of the parent murine antibody LM609.

The invention is additionally directed to optimized forms of LM609 grafted antibodies that exhibit increased binding affinity and specificity compared to the non-mouse parental forms of the LM609 grafted antibody.

WO 98/33919 PCT/US98/01826 9 In one embodiment, the hybridoma expressing LM609 was used as a source to generate and clone cDNAs encoding LM609. The heavy and light chain encoding cDNAs were sequenced and their CDR regions were substituted into a human antibody framework to generate the non-murine form of the antibody. The substitution or grafting of the CDRs was performed by codon-based mutagenesis to generate a combinatorial antibody Fab library consisting of members that presented alternative residues at certain positions. Screening of the library resulted in the isolation of Vitaxin. As a grafted antibody containing human framework sequences, it is unlikely that Vitaxin will elicit a host immune response and can therefore be advantageously used for the treatment of ayP3-mediated diseases.

As used herein, the term "monoclonal antibody LM609" or "LM609" is intended to mean the murine monoclonal antibody specific for the integrin 0a03 which is described by Cheresh, D.A. Proc. Natl. Acad. Sci. USA 84:6471-6475 (1987) and by Cheresh and Spiro J. Biol.

Chem. 262:17703-17711 (1987). LM609 was produced against and is reactive with the M21 cell adhesion receptor now known as the integrin 0(3. LM609 inhibits the attachment of M21 cells to 0 A 3 ligands such as vitronectin, fibrinogen and von Willebrand factor (Cheresh and Spiro, supra) and is also an inhibitor of aN3-mediated pathologies such as tumor induced angiogenesis (Brooks et al. Cell 79:1157-1164 (1994), granulation tissue development in cutaneous wound (Clark et al., Am. J.

Pathology, 148:1407-1421 (1996)) and smooth muscle cell migration such as that occurring during restenosis (Choi et al., J. Vascular Sura., 19:125-134 (1994); Jones et al., Proc. Natl. Acad. Sci. 93:2482-2487 (1996)).

WO 98/33919 PCT/US98/01826 As used herein, the term "Vitaxin" is intended to refer to a non-mouse antibody or functional fragment thereof having substantially the same heavy and light chain CDR amino acid sequences as found in LM609. The term "Vitaxin" when used in reference to heavy or light chain polypeptides is intended to refer to a non-mouse heavy or light chain or functional fragment thereof having substantially the same heavy or light chain CDR amino acid sequences as found in the heavy or light chain of LM609, respectively. When used in reference to a functional fragment, not all LM609 CDRs need to be represented. Rather, only those CDRs that would normally be present in the antibody portion that corresponds to the functional fragment are intended to be referenced as the LM609 CDR amino acid sequences in the Vitaxin functional fragment. Similarly, the use of the term "Vitaxin" in reference to an encoding nucleic acid is intended to refer to a nucleic acid encoding a non-mouse antibody or functional fragment having substantially the same nucleotide sequence as the heavy and light chain CDR nucleotide sequences and encoding substantially the same CDR amino acid sequences as found in LM609.

As used herein, the term "LM609 grafted antibody" is intended to refer to a non-mouse antibody or functional fragment thereof having substantially the same heavy and light chain CDR amino acid sequences as found in LM609 and absent of the substitution of LM609 amino acid residues outside of the CDRs as defined by Kabat et al., U.S. Dept. of Health and Human Services, "Sequences of Proteins of Immunological Interest" (1983). The term "LM609 grafted antibody" or "LM609 grafted" when used in reference to heavy or light chain polypeptides is intended to refer to a non-mouse heavy or light chain or functional fragment thereof having substantially the same WO 98/33919 PCT/US98/01826 11 heavy or light chain CDR amino acid sequences as found in the heavy or light chain of LM609, respectively, and also absent of the substitution of LM609 residues outside of the CDRs as defined by Kabat et al., supra. When used in reference to a functional fragment, not all LM609 CDRs need to be represented. Rather, only those CDRs that would normally be present in the antibody portion that corresponds to the functional fragment are intended to be referenced as the LM609 CDR amino acid sequences in the LM609 grafted functional fragment. Similarly, the term "LM609 grafted antibody" or "LM609 grafted" used in reference to an encoding nucleic acid is intended to refer to a nucleic acid encoding a non-mouse antibody or functional fragment being absent of the substitution of LM609 amino acids outside of the CDRs as defined by Kabat et al., supra and having substantially the same nucleotide sequence as the heavy and light chain CDR nucleotide sequences and encoding substantially the same CDR amino acid sequences as found in LM609 and as defined by Kabat et al., supra.

The term "grafted antibody" or "grafted" when used in reference to heavy or light chain polypeptides or functional fragments thereof is intended to refer to a heavy or light chain or functional fragment thereof having substantially the same heavy or light chain CDR of a donor antibody, respectively, and also absent of the substitution of donor amino acid residues outside of the CDRs as defined by Kabat et al., supra. When used in reference to a functional fragment, not all donor CDRs need to be represented. Rather, only those CDRs that would normally be present in the antibody portion that corresponds to the functional fragment are intended to be referenced as the donor CDR amino acid sequences in the functional fragment. Similarly, the term "grafted WO 98/33919 PCT/US98/01826 12 antibody" or "grafted" when used in reference to an encoding nucleic acid is intended to refer to a nucleic acid encoding an antibody or functional fragment, being absent of the substitution of donor amino acids outside of the CDRs as defined by Kabat et al., supra and having substantially the same nucleotide sequence as the heavy and light chain CDR nucleotide sequences and encoding substantially the same CDR amino acid sequences as found in the donor antibody and as defined by Kabat et al., supra.

The meaning of the above terms are intended to include minor variations and modifications of the antibody so long as its function remains uncompromised.

Functional fragments such as Fab, F(ab) 2 Fv, single chain Fv (scFv) and the like are similarly included within the definition of the terms LM609 and Vitaxin. Such functional fragments are well known to those skilled in the art. Accordingly, the use of these terms in describing functional fragments of LM609 or the Vitaxin antibody are intended to correspond to the definitions well known to those skilled in the art. Such terms are described in, for example, Harlow and Lane, Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, New York (1989); Molec. Biology and Biotechnology: A Comprehensive Desk Reference (Myers, R.A. New York: VCH Publisher, Inc.); Huston et al., Cell Biophysics, 22:189-224 (1993); Pluckthun and Skerra, Meth. Enzymol., 178:497-515 (1989) and in Day, E.D., Advanced Immunochemistry, Second Ed., Wiley-Liss, Inc., New York, NY (1990).

As with the above terms used for describing functional fragments of LM609, Vitaxin and a LM609 grafted antibody, the use of terms which reference other WO 98/33919 PCT/US98/01826 13 LM609, Vitaxin or LM609 grafted antibody domains, functional fragments, regions, nucleotide and amino acid sequences and polypeptides or peptides, is similarly intended to fall within the scope of the meaning of each term as it is known and used within the art. Such terms include, for example, "heavy chain polypeptide" or "heavy chain", "light chain polypeptide" or "light chain", "heavy chain variable region" and "light chain variable region" as well as the term "complementarity determining region" (CDR).

In the case where there are two or more definitions of a term which is used and/or accepted within the art, the definition of the term as used herein is intended to include all such meanings unless explicitly stated to the contrary. A specific example is the use of the term "CDR" to describe the non-contiguous antigen combining sites found within the variable region of both heavy and light chain polypeptides. This particular region has been described by Kabat et al., supra, and by Chothia et al., J. Mol. Biol. 196:901-917 (1987) and by MacCallum et al., J. Mol. Biol. 262:732-745 (1996) where the definitions include overlapping or subsets of amino acid residues when compared against each other. Nevertheless, application of either definition to refer to a CDR of LM609, Vitaxin, LM609 grafted antibodies or variants thereof is intended to be within the scope of the term as defined and used herein. The amino acid residues which encompass the CDRs as defined by each of the above cited references are set forth below in Table 1 as a comparison.

14 Table 1: CDR Definitions Kabat' Chothia 2 MacCallum 3 VH CDR1 31-35 26-32 30-35 VH CDR2 50-65 53-55 47-58 VH CDR3 95-102 96-101 93-101 VL CDR1 24-34 26-32 30-36 VL CDR2 50-56 50-52 46-55 V, CDR3 89-97 91-96 89-96 SResidue numbering follows the nomenclature of Kabat et al., supra 2 Residue numbering follows the nomenclature of Chothia et al., supra 3 Residue numbering follows the nomenclature of MacCallum o* et al., supra S• As used herein, the term "substantially" or "substantially the same" when used in reference to a nucleotide or amino acid sequence is intended to mean that the nucleotide or amino acid sequence shows a considerable degree, amount or extent of sequence identity when compared to a reference sequence. Such considerable degree, amount or extent of sequence identity is further considered to be significant and meaningful and therefore exhibit characteristics which are definitively recognizable or known. Thus, a nucleotide sequence which is substantially the same nucleotide sequence as a heavy or light chain of LM609, Vitaxin, or a LM609 grafted antibody including fragments thereof, refers to a sequence which exhibits characteristics that are definitively known or recognizable as encoding or as being the amino acid sequence of LM609, Vitaxin or a LM609 grafted antibody.

Minor modifications thereof are included so long as they \\BRISl\home$\IsabelH\Speci\34980.doc WO 98/33919 PCT/US98/01826 are recognizable as a LM609, Vitaxin or a LM609 grafted antibody sequence. Similarly, an amino acid sequence which is substantially the same amino acid sequence as a heavy or light chain of Vitaxin, a LM609 grafted antibody or functional fragment thereof, refers to a sequence which exhibits characteristics that are definitively known or recognizable as representing the amino acid sequence of Vitaxin or a LM609 grafted antibody and minor modifications thereof.

When determining whether a nucleotide or amino acid sequence is substantially the same as Vitaxin or a LM609 grafted antibody, consideration is given to the number of changes relative to the Vitaxin or LM609 grafted antibody together with whether the function is maintained. For example, a single amino acid change in a 3 amino acid CDR or several changes in a 16 amino acid CDR are considered to be substantially the same if Uvr3 binding function is maintained. Thus, a nucleotide or amino acid sequence is substantially the same if it exhibits characteristics that are definitively known or recognizable as representing the nucleotide or amino acid sequence of Vitaxin or a LM609 grafted antibody and minor modifications thereof as long as Vitaxin or LM609 grafted antibody function is maintained.

As used herein, the term "fragment" when used in reference to a nucleic acid encoding LM609, Vitaxin or a LM609 grafted antibody is intended to mean a nucleic acid having substantially the same sequence as a portion of a nucleic acid encoding LM609, Vitaxin or a LM609 grafted antibody. The nucleic acid fragment is sufficient in length and sequence to selectively hybridize to an LM609, a Vitaxin or a LM609 grafted antibody encoding nucleic acid or a nucleotide sequence 16 that is complementary to an LM609, Vitaxin or LM609 grafted antibody encoding nucleic acid. Therefore, fragment is intended to include primers for sequencing and polymerase chain reaction (PCR) as well as probes for nucleic acid blot or solution hybridization. The meaning of the term is also intended to include regions of nucleotide sequences that do not directly encode LM609 polypeptides such as the introns, and the untranslated region sequences of the LM609 encoding gene.

As used herein, the term "functional fragment" when used in reference to Vitaxin, to a LM609 grafted antibody or to heavy or light chain polypeptides thereof is intended to refer to a portion of Vitaxin or a LM609 grafted antibody including heavy or light chain polypeptides which still retains some or all of the afP 3 binding activity, afP, binding specificity and/or integrin af -inhibitory activity. Such functional fragments can include, for example, antibody functional fragments such as Fab, Fv, single chain Fv (scFv). Other functional fragments can include, for example, heavy or light chain polypeptides, variable region polypeptides or CDR polypeptides or portions thereof so long as such functional fragments retain binding activity, specificity or inhibitory activity. The term is also intended to include polypeptides encompassing, for example, modified forms of naturally occurring amino acids such as D-stereoisomers, non-naturally occurring amino acids, amino acid analogues and mimetics so long as such polypeptides retain functional activity as defined above.

As used herein, the term "enhanced" when used in reference to Vitaxin, a LM609 grafted antibody or a functional fragment thereof is intended to mean that a functional characteristic of the antibody has been \\BRISl\home$\IsabelH\Speci\34980.doc WO 98/33919 PCT/US98/01826 17 altered or augmented compared to a reference antibody so that the antibody exhibits a desirable property or activity. An antibody exhibiting enhanced activity can exhibit, for example, higher affinity or lower affinity binding, or increased or decreased association or dissociation rates compared to a reference antibody. An antibody exhibiting enhanced activity can also exhibit increased stability such as increased half-life in a particular organism. For example, if higher affinity binding is desired, mutations can be introduced into framework or CDR amino acid residues and the resulting antibody variants screened for higher affinity binding to Uv 3 relative to a reference antibody such as the LM609 grafted parent antibody.

The invention provides a nucleic acid encoding a heavy chain polypeptide for Vitaxin or a functional fragment thereof. Also provided is a nucleic acid encoding a light chain polypeptide for Vitaxin or a functional fragment thereof. The nucleic acids consist of substantially the same heavy or light chain variable region nucleotide sequences as those shown in Figure 1A and lB (SEQ ID NOS:l and 3, respectively) or a fragment thereof.

Vitaxin, including functional fragments thereof, is a non-mouse antibody which exhibits substantially the same binding activity, binding specificity and inhibitory activity as LM609. The Vitaxin Fv Fragment was produced by functionally replacing CDRs within human heavy and light chain variable region polypeptides with the CDRs derived from LM609. Functional replacement of the CDRs was performed by recombinant methods known to those skilled in the art.

Such methods are commonly referred to as CDR grafting and WO 98/33919 PCT/US98/01826 18 are the subject matter of U.S. Patent No. 5,225,539.

Such methods can also be found described in "Protein Engineering of Antibody Molecules for Prophylactic and Therapeutic Applications in Man," Clark, M. Nottingham, England: Academic Titles (1993) Briefly, LM609 nucleic acid fragments having substantially the same nucleotide and encoding substantially the same amino acid sequence of each of the heavy and light chain CDRs were synthesized and substituted into each of the respective human chain encoding nucleic acids. To maintain functionality of the newly derived Vitaxin antibody, modifications were performed within the non-CDR framework region. These individual changes were made by generating a population of CDR grafted heavy and light chain variable regions wherein all possible changes outside of the CDRs were represented and then selecting the appropriate antibody by screening the population for binding activity. This screen resulted in the selection of the Vitaxin antibody described herein.

The nucleotide sequences of the Vitaxin heavy and light chain variable regions are shown in Figures 1A and 1B, respectively. These sequences correspond substantially to those that encode the heavy and light chain variable region polypeptides of Vitaxin. These Vitaxin nucleic acids are intended to include both the sense and anti-sense strands of the Vitaxin encoding sequences. Single- and double-stranded nucleic acids are similarly included as well as non-coding portions of the nucleic acid such as introns, and 3'-untranslated regions and regulatory sequences of the gene for example.

WO 98/33919 PCT/US98/01826 19 As shown in Figure 1A, the Vitaxin heavy chain variable region polypeptide is encoded by a nucleic acid of about 351 nucleotides in length which begins at the amino terminal Glnl residue of the variable region through to Serll7. This Vitaxin heavy chain variable region encoding nucleic acid is joined to a human IgGl constant region to yield a coding region of 1431 nucleotides which encodes a heavy chain polypeptide of 477 total amino acids. Shown in Figure 1B is the Vitaxin light chain variable region polypeptide which is encoded by a nucleic acid of about 321 nucleotides in length beginning at the amino terminal Glul residue of the variable region through to Lysl07. This Vitaxin light chain variable region nucleic acid is joined to a human kappa construct region to yield a coding region of 642 nucleotides which code for a light chain polypeptide of 214 total amino acids.

Minor modification of these nucleotide sequences are intended to be included as heavy and light chain Vitaxin encoding nucleic acids and their functional fragments. Such minor modifications include, for example, those which do not change the encoded amino acid sequence due to the degeneracy of the genetic code as well as those which result in only a conservative substitution of the encoded amino acid sequence.

Conservative substitutions of encoded amino acids include, for example, amino acids which belong within the following groups: non-polar amino acids (Gly, Ala, Val, Leu, and Ile); polar neutral amino acids (Cys, Met, Ser, Thr, Asn, and Gin); polar acidic amino acids (Asp and Glu); polar basic amino acids (Lys, Arg and His); and aromatic amino acids (Phe, Trp, Tyr, and His). Other minor modifications are included within the nucleic acids encoding Vitaxin heavy and light WO 98/33919 PCT/US98/01826 chain polypeptides so long as the nucleic acid or encoded polypeptides retain some or all of their function as described herein.

Thus, the invention also provides a nucleic acid encoding a Vitaxin heavy chain or functional fragment thereof wherein the nucleic acid encodes substantially the same heavy chain variable region amino acid sequence of Vitaxin as that shown in Figure 1A (SEQ ID NO:2) or a fragment thereof. Similarly, the invention also provides a nucleic acid encoding a Vitaxin light chain or functional fragment thereof wherein the nucleic acid encodes substantially the same light chain variable region amino acid sequence of Vitaxin as that shown in Figure lB (SEQ ID NO:4) or a fragment thereof.

In addition to conservative substitutions of amino acids, minor modifications of the Vitaxin encoding nucleotide sequences which allow for the functional replacement of amino acids are also intended to be included within the definition of the term. The substitution of functionally equivalent amino acids encoded by the Vitaxin nucleotide sequences is routine and can be accomplished by methods known to those skilled in the art. Briefly, the substitution of functionally equivalent amino acids can be made by identifying the amino acids which are desired to be changed, incorporating the changes into the encoding nucleic acid and then determining the function of the recombinantly expressed and modified Vitaxin polypeptide or polypeptides. Rapid methods for making and screening multiple simultaneous changes are well known within the art and can be used to produce a library of encoding nucleic acids which contain all possible or all desired changes and then expressing and screening the library for WO 98/33919 PCT/US98/01826 21 Vitaxin polypeptides which retain function. Such methods include, for example, codon based mutagenesis, random oligonucleotide synthesis and partially degenerate oligonucleotide synthesis.

Codon based mutagenesis is the subject matter of U.S. Patent Nos. 5,264,563 and 5,523,388 and is advantageous for the above procedures since it allows for the production of essentially any and all desired frequencies of encoded amino acid residues at any and all particular codon positions within an oligonucleotide.

Such desired frequencies include, for example, the truly random incorporation of all twenty amino acids or a specified subset thereof as well as the incorporation of a predetermined bias of one or more particular amino acids so as to incorporate a higher or lower frequency of the biased residues compared to other incorporated amino acid residues. Random oligonucleotide synthesis and partially degenerate oligonucleotide synthesis can similarly be used for producing and screening for functionally equivalent amino acid changes. However, due to the degeneracy of the genetic code, such methods will incorporate redundancies at a desired amino acid position. Random oligonucleotide synthesis is the coupling of all four nucleotides at each nucleotide position within a codon whereas partially degenerate oligonucleotide synthesis is the coupling of equal portions of all four nucleotides at the first two nucleotide positions, for example, and equal portions of two nucleotides at the third position. Both of these latter synthesis methods can be found described in, for example, Cwirla et al., Proc. Natl. Acad. Sci. USA 87:6378-6382, (1990) and Devlin et al., Science 249:404- 406, (1990).

WO 98/33919 PCT/US98/01826 22 Identification of amino acids to be changed can be accomplished by those skilled in the art using current information available regarding the structure and function of antibodies as well as available and current information encompassing methods for CDR grafting procedures. For example, CDRs can be identified within the donor antibody by any or all of the criteria specified in Kabat et al., supra, Chothia et al., supra, and/or MacCallum et al., supra, and any or all non-identical amino acid residues falling outside of these CDR sequences can be changed to functionally equivalent amino acids. Using the above described methods known within the art, any or all of the non-identical amino acids can be changed either alone or in combination with amino acids at different positions to incorporate the desired number of amino acid substitutions at each of the desired positions. The Vitaxin polypeptides containing the desired substituted amino acids are then produced and screened for retention or augmentation of function compared to the unsubstituted Vitaxin polypeptides. Production of the substituted Vitaxin polypeptides can be accomplished by, for example, recombinant expression using methods known to those skilled in the art. Those Vitaxin polypeptides which exhibit retention or augmentation of function compared to unsubstituted Vitaxin are considered to contain minor modifications of the encoding nucleotide sequence which result in the functional replacement of one or more amino acids.

The functional replacement of amino acids is beneficial when producing grafted antibodies having human framework sequences since it allows for the rapid identification of equivalent amino acid residues without the need for structural information or the laborious WO 98/33919 PCT/US98/01826 23 procedures necessary to assess and identify which amino acid residues should be considered for substitution in order to successfully transfer binding function from the donor. Moreover, it eliminates the actual step-wise procedures to change and test the amino acids identified for substitution. Essentially, using the functional replacement approach described above, all non-identical amino acid residues between the donor and the human framework can be identified and substituted with any or all other possible amino acid residues at each non-identical position to produce a population of substituted polypeptides containing all possible or all desired permutations and combinations. The population of substituted polypeptides can then be screened for those substituted polypeptides which retain function. Using -the codon based mutagenesis procedures described above, the generation of a library of substituted amino acid residues and the screening of functionally replaced residues has been used for the rapid production of grafted therapeutic antibodies as well as for the rapid alteration of antibody affinity. Such procedures are exemplified in, for example, Rosok et al., J. Biol. Chem.

271:22611-22618 (1996) and in Glaser et al., J. Immunol.

149:3903-3913 (1992), respectively.

In addition to framework residues, amino acids in one or more CDRs can be functionally replaced to allow identification of a modified LM609 grafted antibody having enhanced activity. Using the methods described above for framework residues, amino acid substitutions can similarly be introduced into one or more CDRs in an LM609 grafted antibody. The modified LM609 grafted antibody can be tested for binding activity to determine whether avf3 binding activity is maintained. The modified LM609 grafted antibody can be further tested to determine WO 98/33919 PCTIUS98/01826 24 if activity has been enhanced. Functional replacement of amino acid residues in one or more CDRs therefore allows the identification of an enhanced LM609 grafted antibody having a desirable property such as enhanced activity.

To generate modified LM609 grafted antibodies and select those with enhanced activity, several approaches can be employed in the selection of the number of residues within a CDR to mutate as well as the number of CDRs within a LM609 grafted antibody to modify. The choice of selection criteria for mutagenesis of CDRs will depend on the need and desired application of the enhanced antibody. For example, one or a few amino acid positions within a single CDR can be modified to contain selected amino acids at that position. Alternatively, the targeted amino acid positions can be modified to contain all possible amino acids at that position. The resultant population of modified antibodies can then be screened for enhanced activity.

The construction of modified LM609 grafted antibody populations can also be made where all amino acids positions within a CDR have been mutated to contain all possible amino acids and where amino acid positions within multiple CDRs have been modified to contain variant amino acid residues. In this way, populations can be constructed which range from 2 to >107 unique members. The larger the population, the more efficient will be the selection of an enhanced LM609 grafted antibody since there will be a larger number of different antibodies within the population. However, a small population of modified LM609 grafted antibodies can be made and successfully used for the selection of enhanced LM609 grafted antibodies. The size of the population of modified LM609 grafted antibodies will depend on the need WO 98/33919 PCT/US98/01826 of a particular application and can be determined by one skilled in the art.

The generation of modified LM609 grafted antibodies can be achieved by introducing amino acid substitutions into one or more CDRs of an LM609 grafted antibody. For example, single amino acid substitutions can be systematically introduced into a CDR by changing a given amino acid in the CDR to any or all amino acids.

Amino acid substitutions can also be introduced into all amino acid positions in one or more of the CDRs or in all of the CDRs, generating a population of modified LM609 grafted antibody variants. This population of modified LM609 grafted antibody variants having single amino acid substitutions can be screened to identify those variants that maintain Uv03 binding activity. The variants having UVf 3 binding activity can be further characterized to identify those variants having enhanced activity. Such a systematic approach to introducing single amino acid substitutions and generating a population of LM609 grafted antibody variants to screen for enhanced LM609 grafted antibodies having high affinity binding to v03 is described in Example VI.

In addition to generating a population of modified LM609 grafted antibody variants, a particular CDR or a particular amino acid in a CDR can be selected to introduce one or more amino acid substitutions. For example, sequence homology or a structural model can be used to identify particular amino acid positions to introduce amino acid substitutions. In this example, only one or a few modified LM609 grafted antibody variants are generated and screened for binding activity to UA3- One of skill in the art will know or can determine whether it is desirable to generate a large WO 98/33919 PCTIUS98/01826 26 population of modified LM609 grafted antibody variants or to generate a limited number of modified LM609 grafted antibody variants to screen and identify an enhanced LM609 grafted antibodyhaving enhanced activity.

In addition to identifying enhanced LM609 grafted antibodies by generating a population of modified LM609 grafted antibodies having single amino acid substitutions in a CDR and screening for enhanced activity, enhanced LM609 grafted antibody variants can also be generated by combining two or more mutations, each known to independently result in enhanced activity, into a single antibody. When there are more than two mutations, an efficient way to identify combinations of mutations which further augment activity is to construct all possible combinations and permutations and then select for those with enhanced activity. For example, two single mutations in one or more CDRs can be combined to generate a new modified LM609 grafted antibody having two CDR mutations and screened to determine if the af3 3 binding activity is increased over that of the single mutants. Similarly, three mutations can be combined and the resulting modified LM609 grafted antibody screened for enhanced binding activity. Using such an approach of combining CDR mutations, a new population of modified LM609 grafted antibody variants can be generated by incorporating all combinations of the single CDR mutations resulting in enhanced activity into new modified LM609 grafted antibody variants and screening to obtain an optimized enhanced LM609 grafted antibody.

An iterative, step-wise approach to identifying an enhanced LM609 grafted antibody is advantageous in that it allows the identification of an antibody having optimal binding activity without the need to generate and WO 98/33919 PCTIUS98/01826 27 screen a large number of modified LM609 grafted antibody variants. For example, using the approach described in Examples VI and VII in which single mutants were identified and combined into a new population of LM609 grafted antibody variants, enhanced LM609 grafted antibodies having higher affinity were identified by generating 2592 unique variants. In contrast, complete randomization of a single eight amino acid residue CDR would require >1010 unique variants. Therefore, such an iterative approach allows identification of enhanced LM609 grafted antibodies having enhanced activity such as high affinity binding by generating a relatively small number of unique modified LM609 grafted antibody variants and screening and identifying those enhanced LM609 grafted antibody variants exhibiting high affinity binding.

An iterative, step-wise approach to identifying enhanced LM609 variants can also be performed using additional steps. Instead of generating all combinations of single amino acid mutations, the single amino acid mutations can be combined in pairs to generate all combinations of double mutants and screened for activity.

Those double mutants having enhanced activity can be combined with any or all single mutants to generate triple mutants that are screened for enhanced binding activity. Each iterative round of generating modified LM609 grafted antibody variants can incorporate additional single mutations, and the resulting modified LM609 grafted antibodies can be screened for enhanced activity. The step-wise generation of LM609 grafted antibody variants can thus be used to identify an optimized LM609 grafted antibody. Additionally, such an iterative approach also allows for the identification of WO 98/33919 PCTIUS98/01826 28 numerous enhanced antibodies which exhibit a range of different, enhanced binding activities.

An optimized LM609 grafted antibody can also be referred to as an LM609-like grafted antibody or an (v0 3 -specific grafted antibody and is recognizable because the antibody or functional fragments thereof retains the functional characteristics of LM609. For example, enhanced LM609 grafted antibody variants, which have a single amino acid substitution and have enhanced activity, can be identified and correlated with a specific amino acid substitution. These amino acid substitutions can be combined to generate a new modified LM609 grafted antibody that is tested for activity. Such a combination of advantageous CDR amino acid substitutions can result in an optimized LM609 grafted antibody with multiple CDRs having at least one amino acid substitution or a single CDR having multiple amino acid substitutions, where the modified LM609 grafted antibody has enhanced activity.

Enhanced LM609 grafted antibodies, particularly those optimized by functional replacement of amino acid residues in the CDRs, have desirable enhanced properties such as increased affinity. For example, an optimized LM609 grafted antibody having increased affinity will have higher affinity than the parent antibody used for introducing functional replacement of amino acids.

SHigher affinity is determined relative to a reference antibody having a similar structure. For example, if the optimized LM609 grafted antibody is an intact antibody containing two heavy chains and two light chains, then higher affinity is determined relative to the intact parent LM609 grafted antibody. Similarly, if the optimized LM609 grafted antibody is an Fab, then higher WO 98/33919 PCTIUS98/01826 29 affinity is determined relative to the Fab of the parent LM609 grafted antibody.

Although it is not necessary to proceed through multiple optimization steps to obtain a high affinity LM609 grafted antibody, in general, the increase in affinity can correlate with the number of modifications within and between CDRs as well as with the number of optimization steps. Therefore, LM609 grafted antibodies will exhibit a variety of ranges. For example, LM609 grafted antibodies having enhanced affinity will have up to about 2-fold higher affinity or greater, generally greater than about 2- to 5-fold higher affinity such as greater than about 4- to 5-fold higher affinity or about to 10-fold higher affinity than the reference antibody. Particularly, a LM609 grafted antibody having enhanced affinity will have greater than about 10- to higher affinity, greater than about higher affinity, or greater than about 100-fold higher affinity than the reference antibody.

As described above, functional replacement of CDR amino acid residues can be used to identify LM609 grafted antibodies exhibiting higher affinity than a parent LM609 grafted antibody. Methods discussed above or below for introducing minor modifications into Vitaxin or LM609 grafted antibody encoding nucleotide sequences can similarly be used to generate a library of modified LM609 grafted antibody variants, including methods such as codon based mutagenesis, random oligonucleotide synthesis and partially degenerate oligonucleotide synthesis. For example, codon based mutagenesis has been used to generate such a library of modified LM609 grafted antibody variants having single amino acid substitutions (see Example VI) WO 98/33919 PCT/US98/01826 After generating a library of modified LM609 grafted antibody variants, the variants can be expressed and screened for binding activity to av 3 Methods well known to those skilled in the art related to determining antibody-antigen interactions are used to screen for modified LM609 grafted antibodies exhibiting binding activity to aCI3 (Harlow and Lane, supra). For example, an ELISA method has been used to screen a library of modified LM609 grafted antibody variants to identify those variants that maintained cxA3 binding activity (see Example VI). Only those modified LM609 grafted antibodies that maintain av3 binding activity are considered for further characterization.

Modified LM609 grafted antibodies having Uvc 3 binding activity can be further characterized to determine which modified LM609 grafted antibody has enhanced activity. The type of assay used to assess enhanced activity depends on the particular desired characteristic. For example, if altered binding activity is desired, then binding assays that allow determination of binding affinity are used. Such assays include binding assays, competition binding assays and surface plasmon resonance as described in Example VI.

Introduction of single amino acid substitutions into CDRs of LM609 grafted antibodies can be used to generate a library of modified LM609 grafted antibodies and screen for binding activity to U03. Those modified LM609 grafted antibodies exhibiting binding activity to aI 3 can then be further characterized to identify enhanced LM609 grafted antibodies exhibiting enhanced activity such as higher binding affinity. For example, using such an approach, a number of enhanced LM609 grafted antibodies having single amino acid substitutions WO 98/33919 PCT/US98/01826 31 were generated using the heavy chain variable region shown in Figure la (SEQ ID NO:2) and the light chain variable region shown in Figure 7 (SEQ ID NO:32), and LM609 grafted antibodies were identified displaying 2 to 13-fold improved affinity over the parent LM609 grafted antibody (see Example VI) Following identification of enhanced LM609 grafted antibodies having a single amino acid substitution, the amino acid mutations can be combined to further enhance activity. Methods discussed above for introducing single amino acid substitutions into CDRs can similarly be applied to combine amino acid substitutions.

For example, a combinatorial library of amino acid mutations that resulted in enhanced Uv3 binding affinity was generated using degenerate oligonucleotides and two site hybridization mutagenesis as described in Example VII. Enhanced LM609 grafted antibodies containing multiple CDR amino acid substitutions were generated using the heavy chain variable region shown in Figure la (SEQ ID NO:2) and the light chain variable region shown in Figure 7 (SEQ ID NO:32), and LM609 grafted antibodies were identified having 20-fold higher affinity to greater than 90-fold higher affinity than the parent LM609 grafted antibody.

In addition to combining CDR amino acid substitutions to generate an enhanced or optimized LM609 grafted antibody, CDR amino acid substitutions can also be combined with framework mutations that contribute desirable properties to a LM609 grafted antibody. Thus, mutations in CDR or framework regions that enhance activity can be combined to further optimize LM609 grafted antibodies.

WO 98/33919 PCT/US98/01826 32 The invention further provides fragments of Vitaxin heavy and light chain encoding nucleic acids wherein such fragments consist substantially of the same nucleotide or amino acid sequence as the variable region of Vitaxin heavy or light chain polypeptides. The variable region of the Vitaxin heavy chain polypeptide consists essentially of nucleotides 1-351 and of amino acid residues Glnl to Serll7 of Figure 1A (SEQ ID NOS:1 and 2, respectively). The variable region of the Vitaxin light chain polypeptide consists essentially of nucleotides 1-321 and of amino acid residues Glul to Lysl07 of Figure 1B (SEQ ID NOS:3 and 4, respectively).

The termini of such variable region encoding nucleic acids is not critical so long as the intended purpose and function remains the same.

Fragments additional to the variable region nucleic acid fragments are provided as well. Such fragments include, for example, nucleic acids consisting substantially of the same nucleotide sequence as a CDR of a Vitaxin heavy or light chain polypeptide. Sequences corresponding to the Vitaxin CDRs include, for example, those regions defined by Kabat et al., supra, and/or those regions defined by Chothia et al., supra, as well as those defined by MacCallum et al., supra. The Vitaxin CDR fragments for each of the above definitions correspond to the nucleotides set forth below in Table 2.

The nucleotide sequence numbering is taken from the primary sequence shown in Figures 1A and 1B (SEQ ID NOS:1 and 3) and conforms to the definitions previously set forth in Table 1.

WO 98/33919 PCT/US98/01826 33 Table 2: Vitaxin CDR Nucleotide Residues VH CDR1 VH CDR2 V, CDR3 VL CDR1 VL CDR2 VL CDR3 Kabat 91-105 148-198 295-318 70-102 148-168 265-291 Chothia 76-96 157-168 298-315 76-96 148-156 271-288 MacCallum 88-105 139-177 289-315 88-108 136-165 265-288 Similarly, the Vitaxin CDR fragments for each of the above definitions correspond to the amino acid residues set forth below in Table 3. The amino acid residue number is taken from the primary sequence shown in Figures 1A and 1B (SEQ ID NOS:2 and 4) and conforms to the definitions previously set forth in Table 1.

Table 3: Vitaxin CDR Amino Acid Residues VH CDR1 VH CDR2 VH CDR3 VL CDR1 VL CDR2 V, CDR3 Kabat Ser31-Ser35 Lys50-Gly66 His99-Tyrl06 Gln24-His34 Tyr50-Ser56 Gln89-Thr97 Chothia Gly26-Tyr32 Ser53-Gly56 Asn00-Alal05 Ser26-His32 Tyr50-Ser52 Ser91-His96 MacCallum Ser30-Ser35 Trp47-Tyr59 Ala97-Alal05 Ser30-Tyr36 Leu46-Ile55 Gln89-His96 Thus, the invention also provides nucleic acid fragments encoding substantially the same amino acid sequence as a CDR of a Vitaxin heavy or light chain polypeptide.

Nucleic acids encoding Vitaxin heavy and light chain polypeptides and fragments thereof are useful for a variety of diagnostic and therapeutic purposes. For WO 98/33919 PCT/US98/01826 34 example, the Vitaxin nucleic acids can be used to produce Vitaxin antibodies and functional fragments thereof having binding specificity and inhibitory activity against the integrin ov3. The antibody and functional fragments thereof can be used for the diagnosis or therapeutic treatment of af3-mediated disease. Vitaxin and functional fragments thereof can be used, for example, to inhibit binding activity or other functional activities of a v 3 that are necessary for progression of an aPA-mediated disease. Other functional activities necessary for progression of Cv03-mediated disease include, for example, the activation of acf3, ac~v-mediated signal transduction and the avc 3 -mediated prevention of apoptosis. Advantageously, however, Vitaxin comprises non-mouse framework amino acid sequences and as such is less antigenic in regard to the induction of a host immune response. The Vitaxin nucleic acids of the inventions can also be used to model functional equivalents of the encoded heavy and light chain polypeptides.

Thus, the invention provides Vitaxin heavy chain and Vitaxin light chain polypeptides or functional fragments thereof. The Vitaxin heavy chain polypeptide exhibits substantially the same amino acid sequence as that shown in Figure 1A (SEQ ID NO:2) or functional fragment thereof whereas the Vitaxin light chain polypeptide exhibits substantially the same amino acid sequence as that shown in Figure 1B (SEQ ID NO:4) or functional fragment thereof. Also provided is a Vitaxin antibody or functional fragment thereof. The antibody is generated from the above heavy and light chain polypeptides or functional fragments thereof and exhibits selective binding affinity to a03.

WO 98/33919 PCT/US98/01826 The invention provides a nucleic acid encoding a heavy chain polypeptide for a LM609 grafted antibody.

Also provided is a nucleic acid encoding a light chain polypeptide for a LM609 grafted antibody. The nucleic acids consist of substantially the same heavy chain variable region nucleotide sequence as that shown in Figure 1A (SEQ ID NO:l) and substantially the same light chain variable region nucleotide sequence as that shown in Figure 7 (SEQ ID NO:31) or a fragment thereof.

LM609 grafted antibodies, including functional fragments thereof, are non-mouse antibodies which exhibit substantially the same binding activity, binding specificity and inhibitory activity as LM609. The LM609 grafted antibody Fv fragments described herein are produced by functionally replacing the CDRs as defined by Kabat et al. hereinafter referred to as "Kabat CDRs," within human heavy and light chain variable region polypeptides with the Kabat CDRs derived from LM609.

Functional replacement of the Kabat CDRs is performed by the CDR grafting methods previously described and which is the subject matter of U.S. Patent No. 5,225,539, supra. Substitution of amino acid residues outside of the Kabat CDRs can additionally be performed to maintain or augment beneficial binding properties so long as such amino acid substitutions do not correspond to a donor amino acid at that particular position. Such substitutions allow for the modulation of binding properties without imparting any mouse sequence characteristics onto the antibody outside of the Kabat CDRs. Although the production of such antibodies is described herein with reference to LM609 grafted antibodies, the substitution of such non-donor amino acids outside of the Kabat CDRs can be utilized for the production of essentially any grafted antibody. The WO 98/33919 PCTIUS98/01826 36 production of LM609 grafted antibodies is described further below in Example V.

The nucleotide sequences of the LM609 grafted antibody heavy and light chain variable regions are shown in Figures 1A and 7, respectively. These sequences correspond substantially to those that encode the heavy and light chain variable region polypeptides of a LM609 grafted antibody. These nucleic acids are intended to include both the sense and anti-sense strands of the LM609 grafted antibody encoding sequences. Single- and double-stranded nucleic acids are similarly included as well as non-coding portions of the nucleic acid such as introns, and 3'-untranslated regions and regulatory sequences of the gene for example.

The nucleotide and amino acid residue boundaries for a LM609 grafted antibody are identical to those previously described for Vitaxin. For example, a LM609 grafted antibody heavy chain variable region polypeptide is encoded by a nucleic acid of about 351 nucleotides in length which begins at the amino terminal Glnl residue of the variable region through to Serll7 (Figure 1A, SEQ ID NOS:l and 2, respectively). The LM609 grafted antibody light chain variable region polypeptide is encoded by a nucleic acid of about 321 nucleotides in length beginning at the amino terminal Glul residue of the variable region through to Lysl07 (Figure 7, SEQ ID NOS:31 and 32, respectively). As with Vitaxin, minor modification of these nucleotide sequences are intended to be included as heavy and light chain variable region encoding nucleic acids and their functional fragments.

Thus, the invention also provides a nucleic acid encoding a LM609 grafted antibody heavy chain WO 98/33919 PCTIUS98/01826 37 wherein the nucleic acid encodes substantially the same heavy chain variable region amino acid sequence as that shown in Figure 1A (SEQ ID NO:2) or fragment thereof.

Similarly, the invention also provides a nucleic acid encoding a LM609 grafted antibody light chain wherein the nucleic acid encodes substantially the same light chain variable region amino acid sequence as that shown in Figure 7 (SEQ ID NO:32) or fragment thereof.

In addition to conservative substitutions of amino acids, minor modifications of the LM609 grafted antibody encoding nucleotide sequences which allow for the functional replacement of amino acids are also intended to be included within the definition of the term. Identification of amino acids to be changed can be accomplished by those skilled in the art using current information available regarding the structure and function of antibodies as well as available and current information encompassing methods for CDR grafting procedures. The substitution of functionally equivalent amino acids encoded by the LM609 grafted antibody nucleotide sequences is routine and can be accomplished by methods known to those skilled in the art. As described previously, such methods include, for example, codon based mutagenesis, random oligonucleotide synthesis and partially degenerate oligonucleotide synthesis and are beneficial when producing grafted antibodies since they allow for the rapid identification of equivalent amino acid residues without the need for structural information.

The invention further provides fragments of LM609 grafted antibody heavy and light chain encoding nucleic acids wherein such fragments consist substantially of the same nucleotide or amino acid WO 98/33919 PCTIUS98/01826 38 sequence as the variable region of a LM609 grafted antibody heavy or light chain polypeptide. As with Vitaxin, the termini of such variable region encoding nucleic acids is not critical so long as the intended purpose and function remains the same.

Fragments additional to the variable region nucleic acid fragments are provided as well and include, for example, nucleic acids consisting substantially of the same nucleotide sequence as a CDR of a LM609 grafted antibody heavy or light chain polypeptide. As with Vitaxin, sequences corresponding to the LM609 grafted antibody CDRs include, for example, those regions defined by Kabat et al., supra, Chothia et al., supra, as well as those defined by MacCallum et al., supra. The LM609 grafted antibody CDR regions will be similar to those described previously for Vitaxin. Moreover, such regions are well known and can be determined by those skilled in the art given the LM609 sequences and teachings provided herein. Thus, the invention also provides nucleic acid fragments encoding substantially the same amino acid sequence as a CDR of a LM609 grafted antibody heavy or light chain polypeptide.

As with Vitaxin, nucleic acids encoding LM609 grafted antibody heavy and light chain polypeptides and fragments thereof are useful for a variety of diagnostic and therapeutic purposes. For example, LM609 grafted antibody encoding nucleic acids can be used to produce recombinant antibodies and functional fragments thereof having binding specificity and inhibitory activity against the integrin act 3 The antibody and functional fragments thereof can be used for the diagnosis or therapeutic treatment of aB 3 -mediated disease. Such diseases and methods of use for anti-oaI3 antibodies have WO 98/33919 PCT/US98/01826 39 been described previously in reference to Vitaxin and are equally applicable to the LM609 grafted antibodies described herein.

Thus, the invention provides LM609 grafted antibody heavy chain and Vitaxin light chain polypeptides or functional fragments thereof. The LM609 grafted antibody heavy chain polypeptide exhibits substantially the same amino acid sequence as that shown in Figure 1A (SEQ ID NO:2) or functional fragment thereof whereas the LM609 grafted antibody light chain polypeptide exhibits substantially the same amino acid sequence as that shown in Figure 7 (SEQ ID NO:32). Also provided is a LM609 grafted antibody or functional fragment thereof. The antibody is generated from the above heavy and light chain polypeptides or functional fragments thereof and exhibits selective binding affinity to The invention provides an enhanced LM609 grafted antibody exhibiting selective binding affinity to ov( 3 The enhanced LM609 grafted antibody contains at least one amino acid substitution in one or more CDRs of a LM609 grafted heavy chain variable region polypeptide or a LM609 grafted light chain variable region polypeptide, wherein the aP 3 binding affinity of the enhanced LM609 grafted antibody is maintained or enhanced.

To identify enhanced LM609 grafted antibodies, a library of modified LM609 grafted antibodies was generated as described above and in Example VI.

Initially, LM609 CDRs were identified and selected to introduce single amino acid substitutions. Utilizing the numbering system of Kabat et al., supra, the CDR residues selected for mutagenesis were VH CDR1 Gly-Phe-Thr-Phe-Ser- WO 98/33919 PCT/US98/01826 Ser-Tyr-Asp-Met-Ser (SEQ ID NO:34) (Gly 2 -Ser 35 VH CDR2 Trp-Val-Ala-Lys-Val-Ser-Ser-Gly-Gly-Gly (SEQ ID NO:36) and Ser-Thr-Tyr-Tyr-Leu-Asp-Thr-Val-Gln-Gly (SEQ ID NO:38) (Trp 4 -Gly 6 VH CDR3 Ala-Arg-His-Asn-Tyr-Gly-Ser- Phe-Ala-Tyr (SEQ ID NO:40) (Ala 3 -Tyr 102

V

1 CDR1 Gln-Ala- Ser-Gln-Ser-Ile-Ser-Asn-His-Leu-His-Trp-Tyr (SEQ ID NO:42) (Gln 24 -Tyr 36 VL CDR2 Leu-Leu-Ile-Arg-Tyr-Arg-Ser- Gln-Ser-Ile-Ser (SEQ ID NO:44) (Leu"-Ser" 6 and VL CDR3 Gln-Gln-Ser-Gly-Ser-Trp-Pro-His-Thr (SEQ ID NO:46) (Gln"-Thr 9 The nucleotide sequences encoding the CDR residues selected for mutagenesis were CDR1 GGA TTC ACC TTC AGT AGC TATGAC ATG TCT (SEQ ID NO:33); V, CDR2 TGG GTC GCA AAA GTT AGT AGT GGT GGT GGT (SEQ ID NO:35) and AGC ACC TAC TAT TTA GAC ACT GTG CAG GGC (SEQ ID NO:37); VH CDR3 GCA AGA CAT AAC TAC GGC AGT TTT GCT TAC (SEQ ID NO:39); VL CDR1 CAG GCC AGC CAA AGT ATT AGC AAC CAC CTA CAC TGG TAT (SEQ ID NO:41); VL CDR2 CTT CTC ATC CGT TAT CGT TCC CAG TCC ATC TCT (SEQ ID NO:43); and VL CDR3 CAA CAG AGT GGC AGC TGG CCT CAC ACG (SEQ ID Single amino acid substitutions can be introduced into the CDRs of an LM609 grafted antibody to generate a population of modified LM609 grafted antibodies. For example, every amino acid in one or more CDRs can be mutated to any or all amino acids to generate a population of modified LM609 grafted antibodies and the population screened for avP3 binding activity. Although this population is generated by mutating amino acids in CDRs, populations can also be constructed where changes are made in the framework region residues or in both the CDRs and the framework. Such mutations in the variable regions can be made separately, in combination, or stepwise. Thus, the invention also provides for an enhanced WO 98/33919 WO 9833919PCT[US98/01826 41 LM609 grafted antibody, where the amino acid substitution is in the CDR or in the framework region.

The invention additionally provides an enhanced LM609 grafted antibody exhibiting enhanced binding affinity. Enhanced LM609 grafted antibodies exhibiting enhanced binding affinity include those containing at least one of the following CDRs having single amino acid substitutions: a Vw ODRi selected from the group consisting of Gly-Thr- Thr-Phe-Ser-Ser-Tyr-Asp-Met-Ser (SEQ ID NO:48), Gly-Phe- Thr-Trp-Ser-Ser-Tyr-Asp-Met-Ser (SEQ ID NO:50) and Gly- Phe-Thr-Phe-Leu-Ser-Tyr-Asp-Met-Ser (SEQ ID NO:52); a V, CDR2 selected from the group consisting of Trp-Val- Ala-Lys-Val-Lys-Ser-Gly-Gly-Gly (SEQ ID NO:54), Ser-Thr- Tyr-Tyr-Pro-Asp-Thr-Val-Gln-Gly (SEQ ID NO:56) and Ser- Thr-Tyr-Tyr-Leu-Asp-Thr-Val-Glu-Gly (SEQ ID NO:58); a VH. CDR3 selected from the group consisting of Ala-Arg- His-Asn-His-Gly-Ser-Phe-Ala-Tyr (SEQ ID NO:60), Ala-Arg- His-Asn-Tyr-Gly-Ser-Tyr-Ala-Tyr (SEQ ID NO: 62), Ala-Arg- His-Asn-Tyr-Gly-Ser-Phe-Asp-Tyr (SEQ ID NO:64), Ala-Arg- His-Asn-Tyr-Gly-Ser-Phe-Tyr-Tyr (SEQ ID NO:66), Ala-Arg- His-Asn-Tyr-Gly-Ser-Phe-Ala-Ser (SEQ ID NO:68), Ala-Arg- His-Asn-Tyr-Gly-Ser-Phe-Ala-Thr (SEQ ID NO:70), Ala-Arg- His-Asn-Tyr-Gly-Ser-Phe-Ala-Asp (SEQ ID NO:72), Ala-Arg- His-Asn-Tyr-Gly-Ser-Phe-Ala-Glu (SEQ ID NO:74), Ala-Arg- His-Asn-Tyr-Gly-Ser-Phe-Ala-Met (SEQ ID NO:76), Ala-Arg- His-Asn-Tyr-Gly-Ser-Phe-Ala-Gly (SEQ ID NO:78) and Ala- Arg-His-Asn-Tyr-Gly-Ser-Phe-Ala-Ala (SEQ ID the V, CDR1 Gln-Ala-Ser-Gln-Ser-Ile-Ser-Asn-Phe-Leu-His- Trp-Tyr (SEQ ID NO: 82) the VL CDR2 Leu-Leu-Ile-Arg-Tyr- Ser-Ser-Gln-Ser-Ile-Ser (SEQ ID NO:84) and a V. CDR3 selected from the group consisting of Gln-Gln- Ser-Asn-Ser-Trp-Pro-His-Thr (SEQ ID NO:86), Gln-Gln-Ser- Thr-Ser-Trp-Pro-His-Thr (SEQ ID NO:88), Gln-Gln-Ser-Gly- WO 98/33919 PCT/US98/01826 42 Ser-Trp-Pro-Leu-Thr (SEQ ID NO:90) and Gln-Gln-Ser-Gly- Ser-Trp-Pro-Gln-Thr (SEQ ID NO:92).

The nucleotide sequences encoding the CDRs having single amino acid substitutions were VH CDR1 GGA ACT ACC TTC AGT AGC TAT GAC ATG TCT (SEQ ID NO:47), GGA TTC ACC TGG AGT AGC TAT GAC ATG TCT (SEQ ID NO:49), and GGA TTC ACC TTC CTG AGC TAT GAC ATG TCT (SEQ ID NO:51); VH CDR2 TGG GTC GCA AAA GTT AAA AGT GGT GGT GGT (SEQ ID NO:53), AGC ACC TAC TAT CCT GAC ACT GTG CAG GGC (SEQ ID NO:55), and AGC ACC TAC TAT TTA GAC ACT GTG GAG GGC (SEQ ID NO:57); V, CDR3 GCA AGA CAT AAC CAT GGC AGT TTT GCT TAC (SEQ ID NO:59), GCA AGA CAT AAC TAC GGC AGT TAT GCT TAC (SEQ ID NO:61), GCA AGA CAT AAC TAC GGC AGT TTT GAT TAC (SEQ ID NO:63), GCA AGA CAT AAC TAC GGC AGT TTT TAT TAC (SEQ ID NO:65), GCA AGA CAT AAC TAC GGC AGT TTT GCT TCT (SEQ ID NO:67), GCA AGA CAT AAC TAC GGC AGT TTT GCT ACT (SEQ ID NO:69), GCA AGA CAT AAC TAC GGC AGT TTT GCT GAT (SEQ ID NO:71), GCA AGA CAT AAC TAC GGC AGT TTT GCT GAG (SEQ ID NO:73), GCA AGA CAT AAC TAC GGC AGT TTT GCT ATG (SEQ ID NO:75), GCA AGA CAT AAC TAC GGC AGT TTT GCT GGG (SEQ ID NO:77), and GCA AGA CAT AAC TAC GGC AGT TTT GCT GCT (SEQ ID NO:79); VL CDR1 CAG GCC AGC CAA AGT ATT AGC AAC TTT CTA CAC TGG TAT (SEQ ID NO:81); VL CDR2 CTT CTC ATC CGT TAT TCT TCC CAG TCC ATC TCT (SEQ ID NO:83); and VL CDR3 CAA CAG AGT AAT AGC TGG CCT CAC ACG (SEQ ID CAA CAG AGT ACT AGC TGG CCT CAC ACG (SEQ ID NO:87), CAA CAG AGT GGC AGC TGG CCT CTG ACG (SEQ ID NO:89) and CAA CAG AGT GGC AGC TGG CCT CAG ACG (SEQ ID NO:91).

Enhanced LM609 grafted antibodies having CDRs with single amino acid substitutions and higher affinity binding than the parent LM609 grafted antibody can also be identified, where the corresponding amino acid mutations are combined to generate new modified LM609 WO 98/33919 PCTIUS98/01826 43 grafted antibodies. Identification is performed by screening for o P 3 binding activity. In some combinations, the LM609 grafted antibody will comprise at least one CDR having two or more amino acid substitutions. The invention provides an enhanced LM609 grafted antibody containing at least one of the following CDRs containing multiple amino acid substitutions: a V, CDR3 selected from the group consisting of Ala-Arg-His- Asn-His-Gly-Ser-Phe-Ala-Ser (SEQ ID NO:94); Ala-Arg-His- Asn-His-Gly-Ser-Phe-Tyr-Ser (SEQ ID NO:96); Ala-Arg-His- Asn-Tyr-Gly-Ser-Phe-Tyr-Glu (SEQ ID NO:98); and Ala-Arg- His-Asn-Tyr-Gly-Ser-Phe-Tyr-Ser (SEQ ID NO:100).

The nucleotide sequences encoding the CDRS having multiple amino acid substitutions were VH CDR3 GCA AGA CAT AAC CAT GGC AGT TTT GCT TCT (SEQ ID NO:93), GCA AGA CAT AAC CAT GGC AGT TTT TAT TCT (SEQ ID NO:95), GCA AGA CAT AAC TAC GGC AGT TTT TAT GAG (SEQ ID NO:97), and GCA AGA CAT AAC TAC GGC AGT TTT TAT TCT (SEQ ID NO:99).

The invention also provides an enhanced LM609 grafted antibody exhibiting selective binding affinity to UvP3, wherein the enhanced LM609 grafted antibody contains at least one amino acid substitution in two or more CDRs of a LM609 grafted heavy chain variable region polypeptide or a LM609 grafted light chain variable region polypeptide.

An enhanced LM609 grafted antibody containing at least one amino acid substitution in two or more CDRs of a LM609 grafted heavy chain variable region polypeptide or a LM609 grafted light chain variable region polypeptide can include an LM609 grafted antibody containing the combination of CDRs selected from the group consisting of: the VL CDR1 SEQ ID NO:57 and the VH WO 98/33919 PCT/US98/01826 44 CDR3 SEQ ID NO:50; the VL CDRl SEQ ID NO:57, the CDR2 SEQ ID NO:44 and the VH CDR3 SEQ ID NO:50; the VL CDRl SEQ ID NO:57, the VH CDR2 SEQ ID NO:44 and the VH CDR3 SEQ ID NO:52; the VL CDR1 SEQ ID NO:57, the V, CDR2 SEQ ID NO:44 and the VH CDR3 SEQ ID NO:51; the VL CDRl SEQ ID NO:57 and the VH CDR3 SEQ ID NO:52; the VL CDR3 SEQ ID NO:59, the V, CDR2 SEQ ID NO:44 and the CDR3 SEQ ID NO:50; the VL CDR3 SEQ ID NO:61 and VH CDR3 SEQ ID NO:50; and the V, CDR3 SEQ ID NO:61, the V, CDR2 SEQ ID NO:44 and VH CDR3 SEQ ID In addition to enhanced LM609 grafted antibodies containing two or more CDRs having single amino acid substitutions, the invention also provides enhanced LM609 grafted antibodies wherein at least one of the CDRs has two or more amino acid substitutions.

Enhanced LM609 grafted antibodies having at least one CDR with two or more amino acid substitutions can include those containing the combination of CDRs selected from the group consisting of: the VL CDR1 SEQ ID NO:57, the VH CDR2 SEQ ID NO:44 and the VH CDR3 SEQ ID NO:63; the VL CDR3 SEQ ID NO:61, the VH CDR2 SEQ ID NO:44 and the VH CDR3 SEQ ID NO:63; the VL CDR3 SEQ ID NO:61, the V, CDR2 SEQ ID NO:44 and the V H CDR3 SEQ ID NO:64; the VL CDR3 SEQ ID NO:61 and the VH CDR3 SEQ ID NO:63; the VL CDR3 SEQ ID NO:61 and the VH CDR3 SEQ ID NO:65; and the V, CDR3 SEQ ID NO:61, the VH CDR2 SEQ ID NO:44 and the V, CDR3 SEQ ID NO:66.

The invention additionally provides a high affinity LM609 grafted antibody exhibiting selective binding affinity to aP 3 The high affinity LM609 grafted antibody contains at least one amino acid substitution in one or more CDRs of a LM609 grafted heavy chain variable WO 98/33919 PCT/US98/01826 region polypeptide or a LM609 grafted light chain variable region polypeptide, wherein the I ,3 binding affinity of the high affinity LM609 grafted antibody is enhanced.

High affinity antibodies can include those containing the combination of CDRs selected from the group consisting of: the VL CDR1 SEQ ID NO:57 and the V, CDR3 SEQ ID NO:50; the VL CDR1 SEQ ID NO:57, the VN CDR2 SEQ ID NO:44 and the VH CDR3 SEQ ID NO:50; the VL CDR1 SEQ ID NO:57, the V, CDR2 SEQ ID NO:44 and the V H CDR3 SEQ ID NO:52; the VL CDR1 SEQ ID NO:57, the VH CDR2 SEQ ID NO:44 and the V, CDR3 SEQ ID NO:51; the VL CDR1 SEQ ID NO:57 and the V, CDR3 SEQ ID NO:52; the VL CDR3 SEQ ID NO:59, the VH CDR2 SEQ ID NO:44 and the VH CDR3 SEQ ID NO:50; the VL CDR3 SEQ ID NO:61, the V, CDR2 SEQ ID NO:44 and the VH CDR3 SEQ ID NO:63; the VL CDR3 SEQ ID NO:61 and VH CDR3 SEQ ID NO:50; the VL CDR3 SEQ ID NO:61, the V, CDR2 SEQ ID NO:44 and V, CDR3 SEQ ID NO:50; the VL CDR1 SEQ ID NO:57, the VH CDR2 SEQ ID NO:44 and the VH CDR3 SEQ ID NO:63; the VL CDR3 SEQ ID NO:61, the VH CDR2 SEQ ID NO:44 and the VH CDR3 SEQ ID NO:64; the VL CDR3 SEQ ID NO:61 and the V, CDR3 SEQ ID NO:63; the VL CDR3 SEQ ID NO:61 and the V

H

CDR3 SEQ ID NO:65; and the VL CDR3 SEQ ID NO:61, the VH CDR2 SEQ ID NO:44 and the V, CDR3 SEQ ID NO:66.

The invention additionally provides a nucleic acid encoding an enhanced LM609 grafted antibody exhibiting selective binding affinity to a,3. The enhanced LM609 grafted antibody encoded by the nucleic acid contains at least one amino acid substitution in one or more CDRs of a LM609 grafted heavy chain variable region polypeptide or a LM609 grafted light chain variable region polypeptide, wherein the va03 binding WO 98/33919 PCT/US98/01826 46 affinity of the enhanced LM609 grafted antibody is maintained or enhanced.

The invention further provides a nucleic acid encoding a high affinity LM609 grafted antibody exhibiting selective binding affinity to av3. The high affinity LM609 grafted antibody encoded by the nucleic acid contains at least one amino acid substitution in one or more CDRs of a LM609 grafted heavy chain variable region polypeptide or a LM609 grafted light chain variable region polypeptide, wherein the a,3 binding affinity of the high affinity LM609 grafted antibody is enhanced.

The invention provides a nucleic acid encoding a heavy chain polypeptide for monoclonal antibody LM609 or functional fragment thereof. Also provided is a nucleic acid encoding a light chain polypeptide for monoclonal antibody LM609 or a functional fragment thereof. The nucleic acids consist of substantially the same heavy or light chain variable region nucleotide sequences as that shown in Figure 2A and 2B (SEQ ID and 7, respectively) or a fragment thereof.

As described previously, monoclonal antibody LM609 has been shown in the art to have binding activity to the integrin ya03. Although specificity can in principle be generated towards essentially any target, LM609 is an integrin inhibitory antibody that exhibits substantial specificity and inhibitory activity to a single member within an integrin family. In this case, LM609 exhibits substantial specificity and inhibitory activity to the ac 3 integrin within the 13 family. The amino acid or nucleotide sequence of monoclonal antibody WO 98/33919 PCT/US98/01826 47 LM609 has never been previously isolated and characterized.

The isolation and characterization of LM609 encoding nucleic acids was performed by techniques known to those skilled in the art and which are described further below in the Examples. Briefly, cDNA from hybridoma LM609 was generated and used as the source for which to isolate LM609 encoding nucleic acids. Isolation was performed by first determining the N-terminal amino acid sequence for each of the heavy and light chain polypeptides and then amplifying by PCR the antibody encoding sequences from the cDNA. The 5' primers were reverse translated to correspond to the newly determined N-terminal amino acid sequences whereas the 3' primers corresponded to sequences substantially similar to antibody constant region sequences. Amplification and cloning of the products resulted in the isolation of the nucleic acids encoding heavy and light chains of LM609.

The nucleotide sequences of the LM609 heavy and light chain variable region sequences are shown in Figure 2A and 2B, respectively. These sequences correspond substantially to those that encode the variable region heavy and light chain polypeptides of LM609. As with the Vitaxin nucleic acids, these LM609 nucleic acids are intended to include both sense and anti-sense strands of the LM609 encoding sequences. Single- and doublestranded nucleic acids are also include as well as non-coding portions of the nucleic acid such as introns, and 3'-untranslated regions and regulatory sequences of the gene for example.

As shown in Figure 2A, the LM609 heavy chain variable region polypeptide is encoded by a nucleic acid WO 98/33919 PCTIUS98/01826 48 of about 351 nucleotides in length which begins at the amino terminal Glul residue of the variable region through to Ala 117. The murine LM609 antibody heavy chain has an IgG2a constant region. Shown in Figure 2B is the LM609 light chain variable region polypeptide which is encoded by a nucleic acid of about 321 nucleotides in length which begins at the amino terminal Aspl residue of the variable region through to Lys 107.

In the functional antibody, LM609 has a kappa light chain constant region.

As with the Vitaxin nucleic acids, minor modifications of these LM609 nucleotide sequences are intended to be included as heavy and light chain LM609 encoding nucleic acids. Such minor modifications are included within the nucleic acids encoding LM609 heavy and light chain polypeptides so long as the nucleic acids or encoded polypeptides retain some or all of their function as described.

Thus, the invention also provides a nucleic acid encoding a LM609 heavy chain or functional fragment wherein the nucleic acid encodes substantially the same variable region amino acid sequence of monoclonal antibody LM609 as that shown in Figure 2A (SEQ ID NO:6) or a fragment thereof. Similarly, the invention also provides a nucleic acid encoding a LM609 light chain or functional fragment wherein the nucleic acid encodes substantially the same variable region amino acid sequence of monoclonal antibody LM609 as that shown in Figure 2B (SEQ ID NO:8) or a fragment thereof.

The invention further provides fragments of LM609 heavy and light chain encoding nucleic acids wherein such fragments consist substantially of the same WO 98/33919 PCT/US98/01826 49 nucleotide or amino acid sequence as the variable region of LM609 heavy or light chain polypeptides. The variable region of the LM609 heavy chain polypeptide consists essentially of nucleotides 1-351 and of amino acid residues Glul to Alall7 of Figure 2A (SEQ ID NOS:5 and 6, respectively). The variable region of the LM609 light chain polypeptide consists essentially of nucleotides 1-321 and of amino acid residues Aspl to Lysl07 of Figure 2B (SEQ ID NOS:7 and 8, respectively). The termini of such variable region encoding nucleic acids is not critical so long as the intended purpose and function remains the same. Such intended purposes and functions include, for example, use for the production of recombinant polypeptides or as hybridization probes for heavy and light chain variable region sequences.

Fragments additional to the variable region nucleic acid fragments are provided as well. Such fragments include, for example, nucleic acids consisting substantially of the same nucleotide sequence as a CDR of a LM609 heavy or light chain polypeptide. Sequences corresponding to the LM609 CDRs include, for example, those regions within the variable region which are defined by Kabat et al., supra, and/or those regions within the variable regions which are defined by Chothia et al., supra, as well as those regions defined by MacCallum et al., supra. The LM609 CDR fragments for each of the above definitions correspond to the nucleotides set forth below in Table 4. The nucleotide sequence numbering is taken from the primary sequence shown in Figures 2A and 2B (SEQ ID NOS:5 and 7) and conforms to the definitions previously set forth in Table WO 98/33919 PCT/US98/01826 Table 4: LM609 CDR Nucleotide Residues VH CDR1 VH CDR2 VH CDR3 VL CDR1 VL CDR2 VL CDR3 Kabat 91-105 148-198 295-318 70-102 148-168 265-291 Chothia 76-96 157-168 298-315 76-96 148-156 271-288 MacCallum 88-105 139-177 288-315 88-108 136-165 265-288 Similarly, the LM609 CDR fragments for each of the above definitions correspond to the amino acid residues set forth below in Table 5. The amino acid residue numbering is taken from the primary sequence shown in Figures 2A and 2B (SEQ ID NOS:6 and 8) and conforms to the definitions set forth in Table 1.

1 1rp~~a1 LM609 CDR Amino Acid Residues .J -J V, CDR1 V, CDR2 VH CDR3 VL CDR1 VL CDR2 VL CDR3 Kabat Ser31-Ser35 Lys50-Gly66 His99-Tyr106 Gln24-His34 Tyr50-Ser56 Gln89-Thr97 Chothia Gly26-Tyr32 Ser53-Gly56 AsnlOO-Alal05 Ser26-His32 Tyr50-Ser52 Ser91-His96 MacCallum Ser30-Ser35 Trp47-Tyr59 Ala97-Alal05 Ser30-Tyr36 Leu46-Ile55 Gln89-His96 Nucleic acids encoding LM609 heavy and light chain polypeptides and fragments thereof are useful for a variety of diagnostic and therapeutic purposes. For example, the LM609 nucleic acids can be used to produce recombinant LM609 antibodies and functional fragments thereof having binding specificity and inhibitory activity against the integrin The antibody and functional fragments thereof can be used to determine the WO 98/33919 PCT/US98/01826 51 presence or absence of a\v 3 in a sample to diagnose the susceptibility or occurrence of an a,03-mediated disease.

Alternatively, the recombinant LM609 antibodies and functional fragments thereof can be used for the therapeutic treatment of a v3_mediated diseases or pathological state. As with Vitaxin, recombinant LM609 and functional fragments thereof can be used to inhibit the binding activity or other functional activities of cP03 that are necessary for progression of the x,P 3 -mediated disease or pathological state.

The LM609 nucleic acids of the invention can also be used to model functional equivalents of the encoded heavy and light chain polypeptides. Such functional equivalents can include, for example, synthetic analogues or mimics of the encoded polypeptides or functional fragments thereof. A specific example would include peptide mimetics of the LM609 CDRs that retain some or substantially the same binding or inhibitory activity of LM609. Additionally, the LM609 encoding nucleic acids can be used to engineer and produce nucleic acids which encode modified forms or derivatives of the antibody LM609, its heavy and light chain polypeptides and functional fragments thereof. As described previously, such modified forms or derivatives include, for example, non-mouse antibodies, their corresponding heavy and light chain polypeptides and functional fragments thereof which exhibit substantially the same binding and inhibitory activity as LM609.

The invention also provides a method of treating an a3-mediated disease. The method consists of administering an effective amount of Vitaxin, a LM609 grafted antibody, an enhanced antibody thereof, or a functional fragment thereof under conditions which allow WO 98/33919 PCT/US98/01826 52 binding to a~33. Also provided is a method of inhibiting a function of av3. The method consists of contacting aov3 with Vitaxin, a LM609 grafted antibody or a functional fragment thereof under conditions which allow binding to a, 3 As described previously, Vitaxin and LM609 grafted antibodies are monoclonal antibodies which exhibit essentially all of the binding characteristics as does its parental CDR-donor antibody LM609. These characteristics include, for example, significant binding specificity and affinity for the integrin ajv 3 The Examples below demonstrate these binding properties and further show that the binding of such antibodies to 0av3 inhibits a03 ligand binding and function. Thus, Vitaxin and LM609 grafted antibodies are useful for a large variety of diagnostic and therapeutic purposes directed to the inhibition of 0a03 function.

The integrin aCp functions in numerous cell adhesion and migration associated events. As such, the dysfunction or dysregulation of this integrin, its function, or of cells expressing this integrin, is associated with a large number of diseases and pathological conditions. The inhibition Ua3 binding or function can therefore be used to treat or reduce the severity of such oa3-mediated pathological conditions.

Described below are examples of several pathological conditions mediated by a, 3 since the inhibition of at least this integrin reduces the severity of the condition. These examples are intended to be representative and as such are not inclusive of all a3 3 -mediated diseases. For example, there are numerous pathological conditions additional to those discussed below which exhibit the dysregulation of a, 3 binding, WO 98/33919 PCT/US98/01826 53 function or the dysregulation of cells expressing this integrin and in which the pathological condition can be reduced, or will be found to be reduced, by inhibiting the binding ca 3 Such pathological conditions which exhibit this criteria, are intended to be included within the definition of the term as used herein.

Angiogenesis, or neovascularization, is the process where new blood vessels form from pre-existing vessels within a tissue. As described further below, this process is mediated by endothelial cells expressing avf3 and inhibition of at least this integrin, inhibits new vessel growth. There are a variety of pathological conditions that require new blood vessel formation or tissue neovascularization and inhibition of this process inhibits the pathological condition. As such, pathological conditions that require neovascularization for growth or maintenance are considered to be a,.3-mediated diseases. The extent of treatment, or reduction in severity, of these diseases will therefore depend on the extent of inhibition of neovascularization.

These a,3-mediated diseases include, for example, inflammatory disorders such as immune and non-immune inflammation, chronic articular rheumatism, psoriasis, disorders associated with inappropriate or inopportune invasion of vessels such as diabetic retinopathy, neovascular glaucoma and capillary proliferation in atherosclerotic plaques as well as cancer disorders.

Such cancer disorders can include, for example, solid tumors, tumor metastasis, angiofibromas, retrolental, fibroplasia, hemangiomas, Kaposi's sarcoma and other cancers which require neovascularization to support tumor growth. Additional diseases which are considered angiogenic include psoriasis and rheumatoid arthritis as well as retinal diseases such as macular degeneration.

WO 98/33919 PCT/US98/01826 54 Diseases other than those requiring new blood vessels which are ac~ 3 -mediated diseases include, for example, 4restenosis and osteoporosis.

Treatment of the c~A 3 -mediated diseases can be performed by administering an effective amount of Vitaxin, a LM609 grafted antibody, an enhanced antibody thereof, or a functional fragment thereof so as to bind to oia and inhibit its function. Administration can be performed using a variety of methods known in the art.

The choice of method will depend on the specific aPI3mediated disease and can include, for example, the in vivo, in situ and ex vivo administration of Vitaxin, a LM609 grafted antibody or functional fragment thereof, to cells, tissues, organs, and organisms. Moreover, such antibodies or functional fragments can be administered to an individual exhibiting or at risk of exhibiting an av(3mediated disease. Definite clinical diagnosis of an ,P 3 mediated disease warrants the administration of Vitaxin, a LM609 grafted antibody or a functional fragment thereof. Prophylactic applications are warranted in diseases where the Ua3-mediated disease mechanisms precede the onset of overt clinical disease. Thus, individuals with familial history of disease and predicted to be at risk by reliable prognostic indicators can be treated prophylactically to interdict a~v-mediated mechanisms prior to their onset.

Vitaxin, a LM609 grafted antibody, an enhanced antibody thereof, or functional fragments thereof can be administered in a variety of formulations and pharmaceutically acceptable media for the effective treatment or reduction in the severity of an aU 3 -mediated disease. Such formulations and pharmaceutically acceptable medias are well known to those skilled in the WO 98/33919 PCT/US98/01826 art. Additionally, Vitaxin, a LM609 grafted antibody or functional fragments thereof can be administered with other compositions which can enhance or supplement the treatment or reduction in severity of an cav3-mediated disease. For example, the coadministration of Vitaxin or a LM609 grafted antibody to inhibit tumor-induced neovascularization and a chemotherapeutic drug to directly inhibit tumor growth is one specific case where the administration of other compositions can enhance or supplement the treatment of an aUv3-mediated disease.

Vitaxin, a LM609 grafted antibody or functional fragments are administered by conventional methods, in dosages which are sufficient to cause the inhibition of avP3 integrin binding at the sight of the pathology.

Inhibition can be measured by a variety of methods known in the art such as in situ immunohistochemistry for the prevalence of a v3 containing cells at the site of the pathology as well as include, for example, the observed reduction in the severity of the symptoms of the aA3-mediated disease.

In vivo modes of administration can include intraperitoneal, intravenous and subcutaneous administration of Vitaxin, a LM609 grafted antibody or a functional fragment thereof. Dosages for antibody therapeutics are known or can be routinely determined by those skilled in the art. For example, such dosages are typically administered so as to achieve a plasma concentration from about 0.01 pg/ml to about 100 pg/ml, preferably about 1-5 pg/ml and more preferably about pg/ml. In terms of amount per body weight, these dosages typically correspond to about 0.1-300 mg/kg, preferably about 0.2-200 mg/kg and more preferably about 0.5-20 mg/kg. Depending on the need, dosages can be WO 98/33919 PCTIUS98/01826 56 administered once or multiple times over the course of the treatment. Generally, the dosage will vary with the age, condition, sex and extent of the 03P-mediated pathology of the subject and should not be so high as to cause adverse side effects. Moreover, dosages can also be modulated by the physician during the course of the treatment to either enhance the treatment or reduce the potential development of side effects. Such procedures are known and routinely performed by those skilled in the art.

The specificity and inhibitory activity of Vitaxin, LM609 grafted antibodies, an enhanced antibody thereof and functional fragments thereof allow for the therapeutic treatment of numerous ctI-mediated diseases.

Such diseases include, for example, pathological conditions requiring neovascularization such as tumor growth, and psoriasis as well as those directly mediated by va,3 such as restenosis and osteoporosis. Thus, the invention provides methods as well as Vitaxin and LM609 grafted antibody containing compositions for the treatment of such diseases.

Throughout this application various publications are referenced within parentheses. The disclosures of these publications in their entireties are hereby incorporated by reference in this application in order to more fully describe the state of the art to which this invention pertains.

It is understood that modifications which do not substantially affect the activity of the various embodiments of this invention are also included within the definition of the invention provided herein.

WO 98/33919 PCT/US98/01826 57 Accordingly, the following examples are intended to illustrate but not limit the present invention.

EXAMPLE I Isolation and Characterization of LM609 Encoding Nucleic Acids This Example shows the cloning and sequence determination of LM609 encoding nucleic acids.

LM609 is directed against the human vitronectin receptor, integrin a,3P. vf 3 is highly upregulated in melanoma, glioblastoma, and mammary carcinoma and plays a role in the proliferation of M21 melanoma cells both in vitro and in vivo. A3 also plays a role in angiogenesis, restenosis and the formation of granulation tissue in cutaneous wounds. LM609 has been shown to inhibit the adhesion of M21 cells to vitronectin as well as prevent proliferation of M21 cells in vitro. Thus, grafting of LM609 could result in a clinically valuable therapeutic agent.

cDNA Synthesis of LM609 Variable Regions: For cDNA synthesis, total RNA was prepared from 108 LM609 hybridoma cells using a modification of the method described by Chomczynski and Sacchi (Chomczynski and Sacchi, Analyt. Biochem. 162:156 (1987)). LM609 variable region genes were cloned by reverse transcription-polymerase chain reaction (RT-PCR) and cDNA was synthesized using BRL Superscript kit. Briefly, 5 ig of total cellular RNA, 650 ng oligo dT and H 2 0 were brought to a total volume of 55 p1. The sample was heated to 70 0 C for 10 min and chilled on ice. Reaction buffer was added and the mixture brought to 10 mM DTT and WO 98/33919 PCT/US98/01826 58 1 mM dNTPs and heated at 37 0 C for 2 minutes. 5 p1 (1000 units) reverse transcriptase was added and incubated at 37°C for 1 hour and then chilled on ice.

All oligonucleotides were synthesized by P-cyanoethyl phosphoramidite chemistry on an ABI 394 DNA synthesizer. Oligonucleotides used for PCR amplification and routine site-directed mutagenesis were purified using oligonucleotide purification cartridges (Applied Biosystems, Foster City, CA). Forward PCR primers were designed from N-terminal protein sequence data generated from purified LM609 antibody. The forward PCR primers contained sequences coding for the first six amino acids in each antibody variable chain (protein sequenced at San Diego State University). The sequence of the light chain forward PCR primer (997) was 5'-GCC CAA CCA GCC ATG GCC GAT ATT GTG CTA ACT CAG-3' (SEQ ID NO:19) whereas the light chain reverse PCR primer (734) was 5'-AC AGT TGG TGC AGC ATC AGC-3' (SEQ ID NO:20) used. This reverse primer corresponds to mouse light chain kappa amino acid residues 109-115. The sequence of the heavy chain forward PCR primer (998) was 5'-ACC CCT GTG GCA AAA GCC GAA GTG CAG CTG GTG GAG-3' (SEQ ID NO:21). Heavy chain reverse PCR primer 733: 5'-GA TGG GGG TGT CGT TTT GGC-3' SEQ ID NO:22). The PCR primers also contain regions of homology with specific sequences within the immunoexpression vector.

VL and VH chains were amplified in two separate 1l reaction mixtures containing 2 il of the cDNA-RNA heteroduplex, 66.6 mM Tris-HCl pH 8.8, 1.5 mM MgCl2, 0.2 mM of each four dNTPs, 10 mM 2-mercaptoethanol, 0.25 units Taq polymerase (Boehringer-Mannheim, Indianapolis, IN) and 50 pmoles each of primers 997 and 734 and 998 and WO 98/33919 PCT/US98/01826 59 733, respectively. The mixtures were overlaid with mineral oil and cycled for two rounds of PCR with each cycle consisting of 30 seconds at 94°C (denature), seconds at 50°C (anneal), and 30 seconds at 72°C (synthesis). This reaction was immediately followed by cycles of PCR consisting of 30 seconds at 94°C (denature), 30 seconds at 55°C (anneal), and 30 seconds at 720C (synthesis) followed by a final synthesis reaction for 5 minutes at 72 0 C. The reaction products were pooled, extracted with CHCl 3 and ethanol precipitated.

Amplified products were resuspended in 20 ±1 TE buffer (10 mM Tris-HC1, 1 mM EDTA, pH 8.0) and electrophoresed on a 5% polyacrylamide gel. Bands migrating at expected molecular weights of V H and VL were excised, chemically eluted from the gel slice, extracted with organic solvents and ethanol precipitated.

Cloning of amplified VH and VL genes into M13 phage immunoexpression vector: The amplified V region gene products were sequentially cloned into the phage immunoexpression vector by hybridization mutagenesis (Near, R. Biotechniques 12:88 (1992); Yelton et al., J.

Immunol. 155:1994-2003 (1995)). Introduction of the amplified VL and VH sequences by hybridization mutagenesis positions the antibody sequences in frame with the regulatory elements contained in the M13 vector required for efficient Fab expression. One advantage of this technique is that no restriction endonuclease sites need to be incorporated into the VL or VH gene sequences for cloning as is done with conventional DNA ligation methods.

To perform the cloning, 400 ng each of the double-stranded amplified products were first WO 98/33919 PCT/US98/01826 phosphorylated with polynucleotide kinase. 100 ng of the phosphorylated LM609 VL product was mixed with 250 ng of uridinylated BS11 phage immunoexpression vector, denatured by heating to 90°C and annealed by gradual cooling to room temperature. BS11 is an M13 immunoexpression vector derived from M13 IX and encodes CHi of murine IgG1 and murine kappa light chain constant domain (Huse, W.D. In: Antibody Engineering: A Practical Guide, C.A.K. Borrebaeck, ed. W.H. Freeman and Co., Publishers, New York, pp. 103-120 (1991)). Nucleotide sequences included in the PCR amplification primers anneal to complementary sequences present in the singlestranded BS11 vector. The annealed mixture was fully converted to a double-stranded molecule with T4 DNA polymerase plus dNTPs and ligated with T4 ligase. 1 pl of the mutagenesis reaction was electroporated into E.

coli strain DH10B, titered onto a lawn of XL-1 E. coli and incubated until plaques formed. Plaque lift assays were performed as described using goat anti-murine kappa chain antibody conjugated to alkaline phosphatase (Yelton et al, supra; Huse, supra). Fifteen murine light chain positive M13 phage clones were isolated, pooled and used to prepare uridinylated vector to serve as template for hybridization mutagenesis with the PCR amplified LM609 VH product.

Clones expressing functional murine LM609 Fab were identified by binding to purified axI3 by ELISA.

Briefly, Immulon II ELISA plates were coated overnight with 1 ig/ml (100 ng/well) aP 3 and nonspecific sites blocked for two hours at 270C. Soluble Fabs were prepared by isolating periplasmic fractions of cultures of E. coli strain MK30-3 (Boehringer Mannheim Co.) infected with the Fab expressing M13 phage clones. Periplasm fractions WO 98/33919 PCT/US98/01826 61 were mixed with binding buffer 100 mM NaC1, 50 mM Tris pH 7.4, 2mM CaC1,, 1 mM MgC2,, 1 mM MnC 2 0.02% NaN 3 1 mg/ml BSA and incubated with immobilized cx 3 for two hours at 27 0 C. Plates were washed with binding buffer and bound Fab detected with goat anti-murine kappa chain antibody conjugated to alkaline phosphatase. Four aP3 reactive clones were identified: muLM609M13 12, 29, 31 and 69.

MuLM609M13 12 and 29 gave the strongest signals in the ELISA assay. DNA sequence analysis showed that clones muLM609M13 12, 31 and 69 all had identical light chain sequence and confirmed the previously determined N-terminal amino acid sequence of purified LM609 light chain polypeptide. All four clones had identical VH DNA sequence and also confirmed the previously determined N-terminal amino acid sequence of purified LM609 heavy chain polypeptide.

To further characterize the binding activity of each clone, soluble Fab fractions were prepared from ml cultures of E. coli strain MK30-3 infected with clones 12 and 29 and evaluated for binding to (a3 in a competitive ELISA with LM609 IgG. The results of this ELISA are shown in Figure 3. Clone muLM609M13 12 was found to inhibit LM609 IgG binding (at LM609 IgG concentrations of 1 ng/ml and 5 ng/ml) to a(4 3 in a concentration dependent manner at periplasm titers ranging from neat to 1:80. Clone muLM609M13 12 was plaque purified and both the V region heavy and light chain DNA sequences again determined. Complete DNA sequence of the final clone, muLM609M13 12-5, is shown in -Figures 2A and 2B.

WO 98/33919 PCT/US98/01826 62 EXAMPLE II Construction of Vitaxin: A CDR Grafted LM609 Functional Fragment One goal of grafting antibodies is to preserve antibody specificity and affinity when substituting non-human CDRs into a human antibody framework. Another goal is to minimize the introduction of foreign amino acid sequences so as to reduce the possible antigenicity with a human host. This Example describes procedures for accomplishing both of these goals by producing libraries of grafted antibodies which represent all possible members which exhibit the highest affinities for the desired antigen.

The above library was constructed in E. coli wherein the possible CDR and framework changes were incorporated using codon-based mutagenesis (Kristensson et al., In: Vaccines 95. Cold Spring Harbor Laboratory Press. Cold Spring Harbor, NY (1995); Rosok et al., J.

Biol. Chem. (271:22611-22613 (1996)). Using these procedures, a library was constructed and a functionally active humanized anti-a 3 -inhibitory antibody was identified.

For the construction of one grafted form of LM609, human framework sequences showing the highest degree of identity to the murine LM609 V region gene sequences were selected for receiving the LM609 CDRs.

Human heavy chain V region M72 'CL (HHC30Q, HC Subgroup 3, Kabat et al., supra) had 88% identity to frameworks 1, 2 and 3 of LM609 heavy chain and human light chain V region LS1 'CL (HKL312, Kappa subgroup 3, Kabat et al., supra) had 79% identity to frameworks 1, 2 and 3 of LM609 WO 98/33919 PCT/US98/01826 63 light chain. Murine LM609 CDR sequences, as defined by Kabat et al., supra were grafted onto the human frameworks. Residues predicted to be buried that might affect the structure and therefore the binding properties of the original murine combining site were taken into consideration when designing possible changes (Singer et al., supra; Padlan, E.A. Mol. Immunol. 28:489-498 (1991)). This analysis of framework residues considered to be important for preserving the specificity and affinity of the combining site revealed only a few differences. For example, in the heavy chain sequence, the predicted buried residues displayed 100% identity.

Of particular note is that Argl6 in human heavy chain V region M72 'CL is a relatively uncommon residue among human chains. However, this residue was also found to be present in LM609 V, and therefore was retained.

Similarly, Argl9 in LM609 is a relatively rare residue among murine heavy chains but it is found to occur in M72 'CL and was therefore retained. In the light chain sequences, two nonidentical buried residues were identified between LM609 and LS1 'CL framework regions at positions 49 and 87. These two positions were therefore incorporated into the grafted antibody library as both human and murine alternatives.

Full-length grafted V region genes were synthesized by PCR using long overlapping oligonucleotides. Light chain oligonucleotides containing mixed amino acid residues at positions 49 and 87 were synthesized as described in Glaser et al. (J.

Immunol. 149:3903-3913 (1992)) and as illustrated in the oligonucleotides represented as VL oligo3 and VL oligo4.

(SEQ ID NOS:16 and 17, respectively). All long oligonucleotides were gel purified.

Graffted L',,609 heavy and light chain V regions were constructed by mixing 5 overlapping oligonucleotides at ecuinolar concentrations, in the presence or9 annealing PCR primer-s. The heav,,y chain oligonucleotides map to the followino nucleotide ocsitions: oligonucleotide 1 oligol) nucleot ides (nt) 1-84; (SEQ 10 NO: 9) oligo2, rt 70-153, (SEQ ID NO: 10) V, oligo3, nt 139-225 (SEQ ID NO:11i); V, oiigo4, nt 211-231 (SEkQ 1.0 NO:121 VH 0lig oS, nt 277-351 (SEQ 10 NO:13) Similarly, the Vitaxi'. light chain oligonucleotides map to the following nucleotide positions: V. oligonucleotide I (VL o1ligo1)0 nucleozides (nt) 1-87; (SEQ ID NO:14); VL oligo2, nt 73-144, (SEQ ID No. 15) VL oligo3, nt 130-213 (SFEQ ID NO: 16) oligo4, nt 199-279 (SEQ ID NO: 17); V oligo5, nt 265-321 (SEQ ID NO:18) The nucleotide-_-:_uenc_2s of oligonucleot ides used to construct grafted LIM6O9 heavy and light chain variable regions are shown in Table 6. Codon positions 49 and 87 in oligo3, and V, oligo4 represent the randomi-zed codons. The annealing primers contained at least 19 nucleotide residues complementary to vector sequences for efficient. annealing of the amplified V region. product to the single-st-randed vect:or. The annealed mixture was fully converted to a double-stranded molecule with T4 DNA polyrnerase plus dNT?s and ligated with T4 ligase.

AMENDED SHEET WO 98/33919 WO 98/39 19PCT/US98O 1826 Table 6: Oligonucleotides-Used to Construct Grafted LM609 Heavy and Light Chain Variable Regions CAGGTGCAGC TGGTGGAGTC TGGGGGAGGC GTTGTGCAGC CTGGAAGGTC CCTGAGACTC TCCTGTGCAG CCTCTGGATT CACC SEQ ID NO: 9 AACTTTTGCG ACCCACTCCA GACCCTTGCC CGGAGCCTGG CGAACCGAAG ACATGTCATA GCTACTGAAG GTGAATCCAG AGGC SEQ ID NO: TGGGTCGCAA AAGTTAGTAG TGGTGGTGGT AGCACCTACT ATTTAGACAC TGTGCAGGGC CGATTCACCA TCTCCAGAGA CAATAGT SEQ ID NO: 11 TGCACAGTAA TACACGGCTG TGTCCTCGGC TCTCAGAGAG TTCATTTGCA GGTATAGGGT GTTCTTACTA TTGTCTCTGG A SEQ ID NO: 12 GTGTATTACT GTGCAAGACA TAACTACGGC AGTTTTGCTT ACTGGGGCCA AGGGACTACA GTGACTGTTT CTAGT SEQ ID NO: 13 GAGATTGTGC TAACTCAGTC TCCAGCCACC CTGTCTCTCA GCCCAGGAGA AAGGGCGACT CTTTCCTGCC AGGCCAGCCA AAGTATT SEQ ID NO: 14 GATGAGAAGC CTTGGGGCTT GACCAGGCCT TTGTTGATAC CAGTGTAGGT GGTTGCTAAT ACTTTGGCTG GC SEQ ID NO: CCAAGGCTTC TCATCWASTA TCGTTCCCAG TCCATCTCTG GGATCCCCGC CAGGTTCAGT GGCAGTGGAT CAGGGACAGA TTTC SEQ ID NO: 16 GCTGCCACTC TGTTGACAGW AATAGACTGC AAAATCTTCA GGCTCCAGAC TGGAGATAGT GAGGGTGAAA TCTGTCCCTG A SEQ ID NO: 17 CAACAGAGTG GCAGCTGGCC TCACACGTTC GGAGGGGGGA CCAAGGTGGA AATTAAG SEQ ID NO: 18 To generate the library, a portion of the mutagenesis reaction (1 4~1) was electroporated into E.

coi strain DH1OB (BRL), titered onto a lawn of XL-l (Stratagene, Inc.) and incubated until plaques formed.

Replica filter lifts were prepared and plaques containing VH gene sequences were screened either by hybridization with a digoxigenin-labeled oligonucleotide complementary to LM609 heavy chain CDR 2 sequences or reactivity with 7F11-alkaline phosphatase conjugate, a monoclonal antibody raised against the decapeptide sequence Tyr Pro Tyr Asp Val Pro Asp Tyr Ala Ser (SEQ ID NO:28) appended to the carboxy terminus of the vector CH, domain (Biosite, WO 98/33919 PCT/US98/01826 66 Inc., San Diego, CA). Fifty clones that were doublepositive were pooled and used to prepare uridinylated template for hybridization mutagenesis with the amplified grafted LM609 VL product.

The mutagenesis reaction was performed as described above with the V, oligonucleotides except that the VL oligonucleotides 1 to 5 were employed (SEQ ID NOS:14 to 18, respectively). The reaction was electroporated into E. coli strain DH10B and filter lifts probed with either goat anti-human kappa chain antibody conjugated to alkaline phosphatase or a goat anti-human Fab antibody using an alkaline phosphatase conjugated rabbit anti-goat secondary reagent for detection.

Positive clones co-expressing both VH and VL gene sequences were selected (160 total) and used to infect E.

coli strain MK30-3 for preparing soluble Fab fragments.

The soluble Fab fragments were screened for binding to av3 in an ELISA assay. Four clones that were shown from the ELISA to strongly bind CvB3 were identified and further characterized. These clones were termed huLM609M13-34, 54, 55 and 145. All four clones were plaque purified and three independent subclones from each clone was used to prepare Fab fragments for additional binding analysis to aP 3 by ELISA.

In this additional ELISA, duplicate plates were coated with aP 3 ligand and incubated with the huLM609 periplasmic samples. In one plate, bound huLM609 Fab was detected with goat anti-human kappa chain antibody conjugated to alkaline phosphatase and in the other plate bound huLM609 Fab was detected with 7F11-alkaline phosphatase conjugate, the monoclonal antibody WO 98/33919 PCT/US98/01826 67 recognizing the decapeptide tag. Subclones huLM609M13- 34-1, 2 and 3 and huLM609M13-145-1, 2 and 3 all yielded double positive signals indicating that the Fabs contain functional VH and VL polypeptides. These results were confirmed in an ELISA assay on M21 cells, a cell line that expresses the integrin (a 3 3 DNA sequence analysis of subclones huLM609M13- 34-3 and huLM609M13-145-3 revealed mutations introduced into the library by errors due to oligonucleotide synthesis or by errors arising during PCR amplification.

These mutations were corrected in clone huLM609M13-34-3 by site-directed mutagenesis. In the light chain sequence the following corrections were made: His36 to Tyr36 and Lysl8 to Argl8. In the heavy chain sequence the following corrections were made: Glul to Glnl, Asn3 to Gln3, Leull to Valll. Additionally, during the construction of LM609 grafted molecules, residue 28 from the heavy chain was considered to be a non-critical framework residue and the human residue (Thr28) was retained. Subsequently, however, it has been determined that residue 28 can be considered part of the CDR.

Therefore, residue 28 was converted to the corresponding mouse residue at that position (Ala28) using site directed mutagenesis with the oligonucleotide 5'-GCT ACT GAA GGC GAA TCC AGA G-3' (SEQ ID NO:29). This change was later determined to not provide benefit over the human framework threonine at this site, and the threonine was retained. The final grafted LM609 clone was designated huLM609M13 1135-4 and is termed herein Vitaxin. The DNA sequence of clone Vitaxin is shown in Figures 2A and 2B.

WO 98/33919 PCT/US98/01826 68 EXAMPLE III Functional Characterization of Vitaxin This Example shows the characterization of Vitaxin's binding specificity, affinity and functional activity in a number of in vitro binding and cell adhesion assays.

The binding specificity of Vitaxin for the integrin aQA was initially assessed by measuring binding to a~0 3 and its crossreactivity to other or 3containing integrins. Specifically, binding specificity was assessed by measuring binding to ouIb 3 the major integrin expressed on platelets, and to oPs, an integrin found prevalent on endothelial cells and connective tissue cell types.

Briefly, to determine crossreactivity, integrins were coated onto an ELISA plate and a series of antibody dilutions were measured for Vitaxin binding activity against a v3 and the other integrins. The integrins 0~V 3 and 0A( 5 were isolated by affinity chromatography as described by Cheresh (1987), supra, and Cheresh and Spiro (1987), supra. a11b, 3 was purchased from CalBiochem. Briefly, an LM609 affinity column (Cheresh and Spiro (1987), supra) was used to isolate U0I 3 from an octylglucoside human placental lysate, whereas an anti-ac affinity column was used to isolate avP5 from the va, 3 -depleted column flow through. Antibody binding activity was assessed by ELISA using a goat anti-human IgG-alkaline phosphatase conjugate. As a control, a purified human IgG 1 antibody was used since Vitaxin contains a human IgG 1 backbone.

WO 98/33919 PCT/US98/01826 69 The results of this assay are shown in Figure 4A and reveal that Vitaxin specifically binds to a~I with high affinity. There was no detectable binding to the other av- or 3 -containing integrins at antibody concentrations over 1.0 mg/ml.

In a further series of binding studies, the binding affinity and specificity was assessed in a competitive binding assay with the parental LM609 antibody against a,3. Competitive binding was measured in an ELISA assay as described above with LM609 being the labeled antibody. Binding of LM609 was determined in the presence of increasing concentrations of Vitaxin competitor. Alternatively, the control competitor antibody was again a human IgGi.

The results of this competition are presented in Figure 4B and show that specific inhibition of LM609 binding can be observed at Vitaxin concentrations of over 0.1 pg/ml. Almost complete inhibition is observed at Vitaxin concentrations greater than 100 pg/ml. This level of competitive inhibition indicates that the parental monoclonal antibody LM609 and the grafted version Vitaxin exhibit essentially identical specificity.

Binding affinity and specificity were also assessed by measuring the inhibitory activity of Vitaxin on v13 3 binding to fibrinogen. For these studies, a,,3 was plated onto ELISA plates as described above for the Vitaxin/al3 binding studies. Inhibitory activity of Vitaxin was determined by measuring the amount of bound biotinylated fibrinogen in the presence of increasing concentrations of Vitaxin or control antibody. Briefly, fibrinogen was purchased from CalBiochem and biotinylated WO 98/33919 PCT/US98/01826 with N-hydroxysuccinimidobiotin as described by the manufacturer (Pierce Life Science and Analytical Research). Streptavidin alkaline phosphatase was used to detect the bound fibrinogen.

The results of this assay are presented in Figure 4C and reveal a specific binding inhibition at Vitaxin concentrations higher than about 0.1 pg/ml.

These results, combined with those presented above showing specific binding of Vitaxin to v3 and competitive inhibition of LM609, demonstrate that Vitaxin maintains essentially all of the binding characteristics and specificity exhibited by the parental murine monoclonal antibody LM609. Described below are additional functional studies which corroborate these conclusions based on in vitro binding assays.

Additional functional studies were performed to further assess the specificity of Vitaxin binding. These studies were directed to the inhibition of integrin a,03 binding in cell adhesion assays. Endothelial cell adhesion events are an important component in the angiogenic process and inhibition of aP 3 is known to reduce the neovascularization of tumors and thereby reduce the rate of tumor growth. The inhibition of av 3 -mediated cell attachment by Vitaxin in these assays is indicative of the inhibitory activity expected when this antibody is used in situ or in vivo.

Briefly, aP 3 -positive M21 melanoma cells grown in RPMI containing 10% FBS were used for these cell binding assays. Cells were released from the culture dish by trypsinization and re-suspended in adhesion buffer at a concentration of 4 x 10 5 cells/ml (see below) Vitaxin, LM609 or purified human IgG, (control antibody), WO 98/33919 PCT/US98/01826 71 were diluted to the desired concentration in 250 pl adhesion buffer (10 mM Hepes, 2 mM MgCl 2 2 mM CaCl 2 0.2 mM MnCl 2 and 1% BSA in Hepes buffered saline at pH 7.4) and added to wells of a 48-well plate precoated with fibrinogen. The fibrinogen was isolated as described above. Each well was coated with 200 kil fibrinogen at a concentration of 10 Mg/ml for 1 hour at 37 0 C. For the assay, an equal volume of cells (250 pl) containing Vitaxin, LM609 or isotype matched control antibody was added to each of the wells, mixed by gentle shaking and incubated for 20 minutes at 37 0 C. Unbound cells were removed by washing with adhesion buffer until no cells remained in control wells coated with BSA alone. Bound cells were visualized by staining with crystal violet which was subsequently extracted with 100 ul acetic acid and quantitated by determining the absorbance of the solubilized dye at 560 nm.

The results of this assay are shown in Figure and reveal that both Vitaxin and parental antibody LM609 inhibit M21 cell adhesion to fibrinogen over the same concentration range. The inhibitory concentration for 50% maximal adhesion was calculated to be about ng/ml. Specificity of Vitaxin was shown by the lack of inhibition observed by the control IgG, antibody.

In addition to the above cell adhesion results, the inhibitory activity of Vitaxin was also tested in an endothelial cell migration assay. In this.regard, the transwell cell migration assay was used to assess the ability of Vitaxin to inhibit endothelial cell migration (Choi et al., J. Vascular Sura., 19:125-134 (1994) and Leavesly et al., J. Cell Biol, 121:163-170 (1993)).

WO 98/33919 PCT/US98/01826 72 Briefly, human umbilical vein endothelial cells in log phase and at low passage number were harvested by gentle trypsinization, washed and resuspended at a concentration of 2 x 106 cells/ml in 37 0 C HBS containing 1% BSA (20 mM HEPES, 150 mM NaCl, 1.8 mM CaC 2 1.8 mM MgCl 2 5 mM KC1, and 5 mM glucose, pH Antibodies (Vitaxin, LM609, and IgGI control) were diluted to pg/ml from stock solutions. Antibodies were added to cells in a 1:1 dilution (final concentration of antibodies 5 pg/ml; final concentration of cells 1 x 106 cells/ml) and incubated on ice for 10 30 minutes.

The cell/antibody suspensions (200 pl to each compartment) were then added to the upper compartments of a Transwell cell culture chamber (Corning Costar), the lower compartments of which had been coated with 0.5 ml of 10 pg/ml vitronectin (in HBS). Vitronectin serves as the chemoattractant for the endothelial cells. The chambers were placed at 37 0 C for 4 hours to allow cell migration to occur.

Visualization of cell migration was performed by first removing the remaining cells in the upper compartment with a cotton swab. Cells that had migrated to the lower side of the insert were stained with crystal violet for 30 minutes, followed by solubilization in acetic acid and the absorbance of the dye was measured at a wavelength of 550 nm. The amount of absorbance is directly proportional to the number of cells that have migrated from the upper to the lower chamber. The results of the assay are presented in Figure 7B. Both Vitaxin and the parental antibody LM609 yielded essentially identical inhibitory results. Specifically, Vitaxin and LM609 inhibited about 60% of the vitronectininduced migration of endothelial cells compared to the IgG, control and to a sample with no inhibitor.

WO 98/33919 PCT/US98/01826 73 EXAMPLE IV Vitaxin-Mediated Inhibition of v3B, In Animal Models This Example describes the inhibition of tumor growth by Vitaxin in two animal models. Tumor growth was inhibited by inhibiting at least aUf3-mediated neovascularization with Vitaxin.

The first model measures angiogenesis in the chick chorioallantoic membrane (CAM). This assay is a well recognized model for in vivo angiogenesis because the neovascularization of whole tissue is occurring.

Specifically, the assay measures growth factor induced angiogenesis of chicken CAM vessels growing toward the growth factor-impregnated filter disk or into the tissue grown on the CAM. Inhibition of neovascularization is based on the amount and extent of new vessel growth or on the growth inhibition of tissue on the CAM. The assay has been described in detail by others and has been used to measure neovascularization as well as the neovascularization of tumor tissue (Ausprunk et al., Am.

J. Pathol., 79:597-618 (1975); Ossonski et al. Cancer Res., 40:2300-2309 (1980); Brooks et al. Science, 264:569-571 (1994a) and Brooks et al. Cell, 79:1157-1164 (1994b).

Briefly, for growth factor induced angiogenesis filter disks are punched from #1 Whatman Qualitative Circles using a skin biopsy punch. Disks are first sterilized by exposure to UV light and then saturated with varying concentrations of TNF-a or HBSS as a negative control (for at least 1 hour) under sterile conditions. Angiogenesis is induced by placing the saturated filter disks on the CAMs.

WO 98/33919 PCT/US98/01826 74 Inhibition of angiogenesis is performed by treating the embryos with various amounts of Vitaxin and controls (antibody or purified human IgG,). The treatments are performed by intravenous injection approximately 24 hours after disk placement. After 48 hours, CAMs are dissected and angiogenesis is scored on a scale of 1-4. HBSS saturated filter disks are used as the negative control, representing angiogenesis that may occur in response to tissue injury in preparing CAMs, and, values for these CAMS are subtracted out as background. Purified human IgG, is used as the negative control for injections since Vitaxin is of the human IgG, subclass. Vitaxin was found to inhibit TNF-a induced angiogenesis in a dose dependent manner. Maximal inhibition occurred with a single dose of Vitaxin at 300 Mg which resulted in greater than 80% inhibition compared to the human IgG, control.

In addition to the above described CAM assay using growth factor-induced neovascularization, additional studies were performed utilizing tumor-induced neovascularization. For these assays, angiogenesis was induced by transplantating of a3-negative tumor fragments into the CAMs. The use of ac13-negative tumor fragments ensures that any inhibition of tumor growth is due to the inhibition of af3-mediated neovascularization by CAM-derived endothelial cells and not to adhesion events mediated by av3 present on the tumor cells.

Inhibition of tumor growth was assessed by placing a single cell suspension of FG (8 x 106 cells, pancreatic carcinoma) and HEp-3 cells (5 x 105 cells, laryngeal carcinoma) onto CAMs in 30 41. One week later, tumors are removed and cut into approximately 50 mg fragments at which time they are placed onto new CAMs.

WO 98/33919 PCT/US98/01826 After 24 hours of this second placement embryos are injected intravenously with Vitaxin or human IgG as a negative control. The tumors are allowed to grow for about 7 days following which they are removed and weighed.

The results of Vitaxin treatment on the neovascularization of tumors is shown in Figure 6A. The data is expressed as a mean change in tumor weight and demonstrate that Vitaxin is able to inhibit the growth of a, -negative tumors such as FG and HEp-3 tumor fragments.

More specifically, there was a mean weight change for Vitaxin treated FG tumor fragments of -5.38 whereas a change of -11.0 was observed for Vitaxin treated HEp-3 tumors. The IgGi controls exhibited positive mean weight changes of 25.29 and 28.5 for the FG and HEp-3 tumor fragments, respectively. These results were obtained following a single intravenous injection.

In a second animal model, the inhibition of Vx2 carcinoma cells in rabbits was used as a measure of Vitaxin's inhibitory effect on tumors. The Vx2 carcinoma is a transplantable carcinoma derived from a Shope virus-induced papilloma. It was first described in 1940 and has since been used extensively in studies on tumor invasion, tumor-host interactions and angiogenesis. The Vx2 carcinoma is fibrotic in nature, highly aggressive, and exhibits features of an anaplastic type carcinoma.

Propagation of Vx2 tumor is accomplished through serial transplantation in donor rabbits. Following subcutaneous transplantation, it has been reported that after an initial inflammatory reaction, host repair mechanisms set in between days 2 and 4. This repair mechanism is characterized by the formation of new connective tissue and the production of new capillaries. The newly formed WO 98/33919 PCTUS98/01826 76 capillaries are restricted to the repair zone at day 4, however, by day 8 they have extended to the outer region of the tumor. These characteristics and the pharmacokinetics of Vitaxin in rabbits were used to determine initial doses and scheduling of treatments for these experiments. The elimination half life of Vitaxin in animal serum dosed at 1, 5, and 10 mg/kg was found to be 38.9, 60.3, and 52.1 hours, respectively.

Growth of Vx2 tumors in the above animal model was used to study the effect of Vitaxin after early administration on primary tumor growth in rabbits implanted subcutaneously with Vx2 carcinoma. Briefly, Vx2 tumors (50 mg) were transplanted into the inner thigh of rabbits through an incision between the skin and muscle. Measurements of the primary tumor were taken throughout the experiment through day 25. At day 28 after the transplantation animals were sacrificed and tumors were excised and weighed. By day 28, tumors became extremely irregular in shape and as a result, measurements became difficult and were not reflective of tumor volume. Therefore measurements were assessed only through day In a first study, rabbits were treated starting at day 1 post tumor implantation with 5 and 1 mg/kg Vitaxin every four days for 28 days for a total of 7 doses). In both groups, inhibition of tumor growth was observed. In a second series of.studies, rabbits were treated beginning at day 7 post tumor implantation as described above for a total of 5 doses. Inhibition of tumor growth was also observed.

It should be noted that administering a grafted antibody as a repeat dose treatment to rabbits might WO 98/33919 PCT/US98/01826 77 generate an immune response that can have a neutralizing effect on Vitaxin thus potentially comprising efficacy.

Preliminary data suggest that approximately 25-50% of the animals develop such a response.

The results of each of the Vitaxin treatments described above is shown in Figure 6B and 6C. In the rabbits receiving treatments on day 1, inhibition of tumor growth was observed in both the 1 mg/kg and the mg/kg dosing groups compared to the control PBS treated control. Specifically, a growth inhibition of about 67 and 80% was observed, respectively, as measured by the mean tumor weight. A lesser degree of inhibition was observed in animals that began Vitaxin treatment on day 7 post implantation. These results are shown in Figure 6C.

In all cases, inhibition of tumor growth was not see at Vitaxin concentrations lower than 0.2 mg/kg.

EXAMPLE V Construction of LM609 Grafted Functional Antibody Fragments This Example shows the construction of functional LM609 grafted antibody fragments in which only the CDRs have been transferred from the LM609 donor antibody to a human acceptor framework.

CDR grafting of LM609 to produce a functional antibody fragment was accomplished by the methods set forth below. These procedures are applicable for the CDR grafting of essentially any donor antibody where amino acid residues outside of the CDRs from the donor antibody are not desired in the final grafted product.

WO 98/33919 PCTIUS98/01826 78 Briefly, the protein sequence of the LM609 antibody, was determined by cloning and sequencing the cDNA that encodes the variable regions of the heavy and light chains as described in Example I. The CDRs from the LM609 donor antibody were identified and grafted into homologous human variable regions of a human acceptor framework. Identification of CDR regions were based on the combination of definitions published by Kabat et al., and MacCallum et al.

The boundaries of the CDR regions have been cumulatively defined by the above two publications and are residues 30-35, 47-66 and 97-106 for CDRs 1, 2 and 3, respectively, of the heavy chain variable region and residues 24-36, 46-56, and 89-97 for CDRs i, 2 and 3, respectively, of the light chain variable region. Nonidentical donor residues within these boundaries but outside of CDRs as defined by Kabat et al. were identified and were not substituted into the acceptor framework. Instead, functional non-donor amino acid residues were identified and substituted for certain of these non-identical residues.

As described below, the only non-identical residue outside of the CDRs as defined by Kabat et al.

but within the CDRs as defined above is at position 49 of the LM609 light chain. To identify functional non-donor amino acids at this position, a library of nineteen Santibodies was constructed that contained all non-donor amino acids at position 49 and then screened for binding activity against cvI 3 Human immunoglobulin sequences were identified from the Brookhaven Protein Data Bank-Kabat Sequences of Proteins of Immunological Interest database (release WO 98/33919 PCT/US98/01826 79 Human framework sequences showing significant identity to the murine LM609 variable region gene sequences were selected for receiving the LM609 CDRs.

Human heavy chain variable region M72 'CL had 88% identity to frameworks 1, 2 and 3 of LM609 heavy chain and human light chain V region LS1 'CL had 79% identity to frameworks 1, 2 and 3 of LM609 light chain. With the exclusion of non-identical residues outside of the CDRs as defined by Kabat et al. murine LM609 CDR sequences as defined by Kabat et al. and MacCallum et al. were grafted onto the human frameworks. Using this grafting scheme, the final grafted product does not contain any amino acid residues outside of the CDRs as defined by Kabat et al.

which are identical to an LM609 amino acid at the corresponding position (outside of residues: 31-35, 50-66 and 99-106 for CDRs 1, 2 and 3, respectively, of the heavy chain variable region and residues 24-34, 50-56, and 89-97 for CDRs 1, 2 and 3, respectively, of the light chain variable region). Moreover, no intermediates are produced which contain an amino acid residue outside of the CDRs as defined by Kabat et al. which are identical to the LM609 amino acid at that position. The CDR grafting procedures are set forth below.

Full-length CDR grafted variable region genes were synthesized by PCR using long overlapping oligonucleotides as described previously in Example II.

The heavy chain variable region oligonucleotides were Sthose described previously as SEQ ID NOS:9-13. The light chain variable region oligonucleotides were synthesized so as to contain the CDR grafted variable region as well as a stop codon at position 49. The five oligonucleotides for the light chain LM609 grafted variable region are show as SEQ ID NOS:23-27 where the second oligonucleotide in the series contains the stop WO 98/33919 PCT/US98/01826 codon at position 49 (SEQ ID NO:24). The nucleotide sequences of oligonucleotides used to construct LM609 grafted light chain variable region is shown in Table 7.

Table 7: Oligonucleotides Used to Construct LM609 Grafted Light Chain Variable Region GAGATTGTGC TAACTCAGTC TCCAGCCACC CTGTCTCTCA GCCCAGGAGA AAGGGCGACT CTTTCCTGCC AGGCCAGCCA AAGTATT SEQ ID NO: 23 TTAGATGAGA AGCCTTGGGG CTTGACCAGG CCTTTGTTGA TACCAGTGTA GGTGGTTGCT AATACTTTGG CTGGC SEQ ID NO: 24 CCAAGGCTTC TCATCTAATA TCGTTCCCAG TCCATCTCTG GGATCCCCGC CAGGTTCAGT GGCAGTGGAT CAGGGACAGA TTTC SEQ ID NO: GCTGCCACTC TGTTGACAGT AATAGACTGC AAAATCTTCA GGCTCCAGAC TGGAGATAGT GAGGGTGAAA TCTGTCCCTG A SEQ ID NO: 26 CAACAGAGTG GCAGCTGGCC TCACACGTTC GGAGGGGGGA CCAAGGTGGA AATTAAG SEQ ID NO:27 All long oligonucleotides were gel purified.

CDR grafting of the LM609 heavy chain variable region was constructed by mixing 5 overlapping oligonucleotides (SEQ ID NOS:9-13), at equimolar concentrations, in the presence of annealing PCR primers containing at least 18 nucleotide residues complementary to vector sequences for the efficient annealing of the amplified V region product to the single-stranded vector. The annealed mixture was fully converted to a double-stranded molecule with T4 DNA polymerase plus dNTPs and ligated with T4 ligase. The mutagenesis reaction (1 pl) was electroporated into E.

coli strain DH10B (BRL), titered onto a lawn of XL-1 (Stratagene, Inc.) and incubated until plaques formed.

Replica filter lifts were prepared and plaques containing VH gene sequences were screened either by hybridization with a digoxigenin-labeled oligonucleotide complementary to LM609 heavy chain CDR 2 sequences or reactivity with 7F11-alkaline phosphatase conjugate, a monoclonal WO 98/33919 PCT/US98/01826 81 antibody raised against the decapeptide sequence Tyr Pro Tyr Asp Val Pro Asp Tyr Ala Ser (SEQ ID NO:28) appended to the carboxy terminus of the vector CH, domain (Biosite, Inc., San Diego, CA).

Fifty clones that were double-positive were pooled and used to prepare uridinylated template for hybridization mutagenesis with the amplified CDR grafted LM609 VL product constructed in a similar fashion using the five overlapping oligonucleotides shown as SEQ ID NOS:23-27. The mutagenesis reaction was electroporated into E. coli strain DH1OB. Randomly picked clones were sequenced to identify a properly constructed template for construction of the non-donor library at position 49.

This template was prepared as a uridinylated template and an oligonucleotide population of the following sequence was used for site directed mutagenesis.

GGGAACGATA-19aa-GATGAGAAGC The sequence 19aa in the above primer (SEQ ID represents the fact that this primer specifies a sequence population consisting of 19 different codon sequences that encode each of the 19 non-donor amino acids. These amino acids are those not found at position 49 of LM609 and include all amino acids except for Lys. Clones that resulted from this mutagenesis were picked and antibody expressed by these clones were prepared. These samples were then screened for binding to av%3 in an ELISA assay.

Clones having either Arg or Met amino acids in position 49 were functionally identified. The nucleotide and amino acid sequence of the LM609 grafted heavy chain variable region is show in Figure 1A (SEQ ID NOS:1 and 2, respectively). The nucleotide and amino acid sequence of WO 98/33919 PCT/US98/01826 82 the LM609 grafted light chain variable region is shown in Figure 7 (SEQ ID NOS:31 and 32, respectively).

EXAMPLE VI Generation of LM609 Grafted Antibodies Having Enhanced Activity This example shows in vitro maturation of LM609 grafted antibody to obtain antibody variants having increased affinity to avP 3 relative to the parent LM609 grafted antibody.

To optimize the affinity of LM609 grafted antibody in vitro, an M13 phage system was used, which permits the efficient synthesis, expression, and screening of libraries of functional antibody fragments (Fabs). The contribution of each of the six CDRs of the Ig heavy and light chains was assessed. The CDRs were defined broadly based on a combination of sequence variability and antibody structural models (Kabat et al., J. Biol. Chem. 252:6609-6616 (1977); Chothia et al., supra; MacCallum et al., supra). Thus, one library was constructed for each CDR, with the exception of H2 which was split into two libraries due to its long (20 amino acids) length. The variable region frameworks which harbored the mutated CDRs were the heavy chain variable region shown in Figure la (SEQ ID NO:2) and the light chain variable region shown in Figure 7 (SEQ ID NO:32).

CDRs were chosen from the heavy chain variable region shown in Figure la (SEQ ID NO:2) and the light chain variable region shown in Figure 7 (SEQ ID NO:32).

Briefly, utilizing the numbering system of Kabat et al., supra, the residues chosen for mutagenesis of the CDRs WO 98/33919 PCT/US98/01826 83 (Table 9) were: Gln 24 -Tyr 36 in light chain CDR1 (LI); Leu"G-Ser" in light chain CDR2 Gln"-Thr 97 in light chain CDR3 Gly 26 -Ser 3 5 in heavy chain CDR1 (H1); Trp 47 -Gly 65 in heavy chain CDR2 and Ala 93 -Tyr 10 2 in heavy chain CDR3 Libraries were created for each CDR, with the oligonucleotides designed to mutate a single CDR residue in each clone. Due to the extended length of H2, two libraries mutating residues 47-55 (H2a) and 56-65 (H2b), respectively, were constructed to cover this region.

The template for generating light chain CDR3 mutants contained Gly at position 92. However, it was subsequently determined that position 92 of the light chain CDR3 was inadvertently deduced to be a Gly, resulting in humanized LM609 grafted antibodies being constructed with Gly at that position. It was later realized that the original LM609 sequence contained an Asn at position 92. Using the methods described herein to introduce mutations into CDRs of an LM609 grafted antibody, an LM609 grafted antibody having Asn at position 92 of light chain CDR3 was found to have 0Av3 binding activity (see Table confirming the identification of Asn 92 as a functional LM609 grafted antibody. Thus, antibodies containing light chain CDR3 having Gly or Asn at position 92 are active in binding aV 3 Oligonucleotides encoding a single mutation were synthesized by introducing NN(G/T) at each CDR position as described previously (Glaser et al., supra) The antibody libraries were constructed in M131XL604 vector by hybridization mutagenesis as described previously, with some modifications (Rosok et al., J.

Biol. Chem. 271:22611-22618 (1996); Huse et al., J.

84 Immunol. 149:3914-3920 (1992); Kunkel, Proc. Natl. Acad.

Sci. USA 82:488-492 (1985); Kunkel et al., Methods Enzvmol. 154:367-382 (1987)). Briefly, the oligonucleotides were annealed at a 20:1 molar ratio to uridinylated LM609 grafted antibody template (from which the corresponding CDR had been deleted) by denaturing at 0 C for 5 min, ramping to 55'C for 1 h, holding at 55 0

C

for 5 min, then chilling on ice. The reaction was extended by polymerization and electroporated into DH1OB and titered onto a lawn of XL-1 Blue. The libraries consisted of pools of variants, each clone containing a single amino acid alteration in one of the CDR positions.

Utilizing codon-based mutagenesis, every position in all of the CDRs was mutated, one at a time, resulting in the subsequent expression of all twenty amino acids at each CDR residue (Glaser et al., supra). The CDR libraries ranged in size from 288 (L3) to 416 (Ll) unique members and contained a total of 2336 variants.

To permit the efficient screening of the initial libraries, a highly sensitive plaque lift assay, termed capture lift, was employed (Watkins et al., Anal.

Biochem. 256 (1998)). Briefly, phage expression libraries expressing LM609 grafted antibody variants were .4.

initially screened by a modified plaque lift approach, in which the nitrocellulose was pre-coated with goat anti-human kappa antibody and blocked with bovine serum albumin prior to application to the phage-infected bacterial lawn. Following the capture of phage-expressed LM609 grafted antibody variant Fabs, filters were incubated with 1.0 pg/ml biotinylated afJ, for 3 h at 4 0

C,

washed four times, incubated with 2.3 pg/ml NeutrAvidin-alkaline phosphatase (Pierce Chemical Co.; Rockford, IL) for 15 min at 25 0 C, and washed four times.

All dilutions and washes were in binding buffer.

\\BRISl\home$\IsabelH\Speci\34980.doc WO 98/33919 PCT/US98/01826 Variants that bound ovP 3 were identified by incubating the filters for 10-15 min in 0.1 M Tris, pH 9.5, containing 0.4 mM 2,2'-di-p-nitrophenyl-5,5'-diphenyl-3, 3'-(3,3'-dimethoxy-4,4'-diphenylene)ditetrazolium chloride and 0.38 mM 5-bromo-4-chloro-3-indoxyl phosphate mono-(p-toluidinium) salt (JBL Scientific, Inc.; San Luis Obispo, CA).

To generate biotinylated 0, 3 the a0v3 receptor was purified from human placenta by affinity chromatography, as described previously (Smith and Cheresh, J. Biol. Chem. 263:18726-18731 (1988)). To biotinylate a0P3, purified receptor was dialyzed into mM HEPES, pH 7.4, 150 mM NaC1, 1.0 mM CaC 2 containing 0.1% NP-40 (binding buffer) and incubated with 100-fold molar excess sulfosuccinimidobiotin for 3h at 4°C. The reaction was terminated by the addition of 50 mM ethanolamine.

Phage expressed LM609 grafted antibody variants were selectively captured on nitrocellulose filters coated with goat anti-human kappa chain antibody, probed with biotinylated afv3, and detected with NeutrAvidin-alkaline phosphatase. Initially, biotinylated vP3 was titrated on lifts containing phage expressing the LM609 grafted antibody parent molecule only. Subsequently, the concentration of biotinylated oafv was decreased to yield a barely perceptible signal.

In this way, only clones expressing higher affinity variants were readily identified during screening of the variant libraries. Following the exhaustive capture lift screening of 2500 clones from each library, 300 higher affinity variants were identified (see Table 10). The greatest number of clones displaying improved affinity were identified in the H3 (185) and L3 (52) CDRs, though WO 98/33919 PCT/US98/01826 86 variants with improved affinity were identified in every

CDR.

LM609 grafted antibody variants identified by capture lift as having av 3 binding activity were further characterized to determine binding affinity to avx 3 specificity for 0(V3 over other integrins, and a 03 association and dissociation rates. For these assays, purified Fab of LM609 grafted antibody variants was used.

Briefly, Fab was expressed as described previously and was released from the periplasmic space by sonic oscillation (Watkins et al., supra, 1997). Cells collected from one liter cultures were lysed in 10 ml mM Tris, pH 8.0, containing 0.05% Tween 20. Fab was bound to a 1 ml protein A column (Pharmacia) which had been equilibrated with 50 mM glycine, pH 8, containing 250 mM NaCl, washed with the same buffer, and eluted with ml of 100 mM glycine, pH 3, into one-tenth volume 1 M Tris, pH 8. Purified Fab was quantitated as described previously (Watkins et al., supra, 1997).

LM609 grafted antibody variants were tested for binding to IP3 and specificity of binding to 003 relative to avP 5 and a Ib,3. For ELISA titration of Fab on immobilized avA3 and the related integrins vPs, and alIb P3, Immulon II microtiter plates were coated with 1 gg/ml purified receptor in 20 mM Tris, pH 7.4, 150 mM NaCI, 2 mM CaCl 2 1 mM MgCl 2 1 mM MnC1 2 washed once, and blocked in 3% BSA in 50 mM Tris, pH 7.4, 100 mM NaCl, 2 mM CaC1 2 1 mM MgC1 2 for 1 h at 250C. Human aoib3, purified from platelets, was obtained from Enzyme Research Laboratories, Inc. (South Bend, IN) and 0cx, was purified from placental extract depleted of aVP3, as described previously (Smith et al., J. Biol. Chem. 265:11008-11013 (1990)). Just prior to use, the plates were washed two times and were then incubated 1 h at 25 0 C with various dilutions of Fab. The plates were washed five times, incubated 1 h at 25 0 C with goat anti-human kappa-alkaline phosphatase diluted 2000-fold, washed five times, and developed as described previously (Watkins et al., supra, 1997). All dilutions and washes were in 50 mM Tris-HCl, pH 7.4, 100 mM NaCl, 2 mM CaCl,, and 1 mM MgC1,.

Table 8: Capture Lift Screening of LM609 grafted antibody CDR Libraries.

Enhanced Library Size' Screened 2 Positives 3 Affinity 4 H1 320 2500 16 8 H2a 320 5000 26 7 H2b 320 5000 2 1 H3 320 5000 185 785 L1 416 2500 12 1 L2 352 3250 7 1 L3 288 5000 52 41 0 0 0* *r S S S* *r 5 0r.

a

S

0 Number of unique clones based on DNA sequence.

two codons are used to express all twenty amino each position.

Thirtyacids at 2 Phage-expressed libraries were plated on XL-1 Blue/agar lawns at 500-100 plaques per 100 mm dish.

Positives are defined as clones that were identified in the initial screen, replated, and verified in a second capture lift assay.

o.',pUSr \\BRISl\home$\IsabelH\Speci\34980.doc 87a 4Soluble Fab was titrated against immobilized cfit in an ELISA format. Based on comparison of the inflection '0 0 0 0 0 0000 0 0 0000 00 0 00 0 00

P,'

OP S \\BRISl\home$\IsabelH\Speci\34980 .doc WO 98/33919 PCT/US98/01826 88 point of the titration profiles, clones which displayed -3-fold enhanced affinity were selected for further characterization.

0f the 185 positive clones identified by capture lift, 98 were further characterized for binding to immobilized avP3- Figure 8 shows titration of antibody variants and LM609 grafted antibody Fab on immobilized (vA 3 Bacterial cell lysates containing LM609 grafted antibody (closed circles), variants with improved affinity isolated from the primary libraries (S102, closed squares; Y100, open squares; and Y101, open triangles) or from the combinatorial libraries (closed triangles), or an irrelevant Fab (open circles) were titrated on immobilized ov 3 Comparison of the inflection points of the binding profiles obtained from titrating variants on immobilized af3 demonstrated that multiple clones displayed >3-fold improved affinity, confirming the effectiveness of utilizing the capture lift in a semi-quantitative fashion (Figure 8, compare squares and open triangles with closed circles). Based on the capture lift screening and subsequent characterization of binding to immobilized Av 3, it was concluded that both heavy and light chain CDRs are directly involved in the interaction of XVP3 with the LM609 grafted antibody 0 variants.

DNA was isolated from clones displaying >3-fold enhanced binding and sequenced to identify the mutations which resulted in higher affinity. DNA sequencing was performed on isolated single-stranded DNA. The heavy and light chain variable region genes were sequenced by the WO 98/33919 PCT/US98/01826 89 fluorescent dideoxynucleotide termination method (Perkin-Elmer; Foster City, CA). Based on sequence analysis of 103 variants, 23 unique mutations clustered at 14 sites were identified (Table The majority of the sites of beneficial mutations were found in the heavy chain CDRs, with four located in H3, and three each in H2 (2a and 2b combined) and H1. Seven distinct and beneficial amino acid substitutions were identified at a single site within H3, tyrosine residue 102. The diverse nature of the substitutions at this site suggests that tyrosine residue 102 may sterically hinder LM609 grafted antibody binding to aAf 3 In support of this, variants expressing the other aromatic amino acids (phenylalanine, histidine, and tryptophan) instead of tyrosine at residue 102 were never isolated following screening for enhanced binding.

The affinities of select variants were further characterized by utilizing surface plasmon resonance (BIAcore) to measure the association and dissociation rates of purified Fab with immobilized av3. Briefly, surface plasmon resonance (BIAcore; Pharmacia) was used to determine the kinetic constants for the interaction between 0 3 and LM609 grafted antibody variants.

Purified 0Iv3 receptor was immobilized to a (l-ethyl-3-(3-dimethylaminopropyl)-carbodiimide hydrochloride)/N-hydroxysuccinimide-activated sensor chip by injecting 30 4l of 15 Mg/ml c0v3 in 10 mM sodium acetate, pH 4. To obtain association rate constants the binding rate at five different Fab concentrations, ranging from 5-40 g/ml in 50 mM Tris-HCl, pH 7.4, 100 mM NaCl, 2 mM CaC1 2 and 1 mM MgC12, was determined at a flow rate of 10 ul/min. Dissociation rate constants (koff) were the average of five measurements obtained by analyzing the dissociation phase WO 98/33919 PCT/US98/01826 at an increased flow rate (40 l/min). Sensorgrams were analyzed with the BIAevaluation 2.1 program (Pharmacia).

Residual Fab was removed after each measurement with mM HC1, 2 mM CaC1 2 and 1 mM MgC1,.

Table 9 shows that the variants all displayed a lower Kd than the LM609 grafted antibody parent molecule, consistent with both the capture lift and the EL1SA.

Analysis of association and dissociation rates revealed that the majority of improved variants had slower dissociation rates while having similar association rates. For example, LM609 grafted antibody had an association rate 18.0 x 104 M-Is-, while the variants ranged from 16.7-31.8 x 104 M-s l In contrast, every clone dissociated slower than LM609 grafted antibody (4.97 x 10- 3 s 1 with dissociation rates ranging from 1.6-fold (3.03 x 10- 3 s- 1 to 11.8-fold (0.42 x 10- 3 slower.

These results demonstrate that introducing single amino acid substitutions into LM609 grafted antibody CDRs allows the identification of modified LM609 grafted antibodies having higher affinity for av3P than the parent LM609 grafted antibody.

Table 9 detf2ain fEhacd M0 Grafted AIbodie foPrimary Libraries Chain libaryt uL k (X10' xO' K M 6 CD1

GFTF

T27 n~d d.n d.

W29 wn~d.n~d.n.d.

Lnd d n.d.

CAH CDR2a W VA K VSS G GG 1782.18 1.

U) 5 CDR2b S T Y Y L 1) T V Q G3.815 P3181 5 5.8 4 4En. d. n. d. n. d.

C

D 14. CD 3A R H N m n .d d S1224.2 1.44 T12C 24.6 1.43 5.8 rl0 27.6 0.97 E] D02 ED .nd id 102 En.d. n.d. nd.

G102 G 16.1 2.01 12.5 A102 A27.5 2.27 8.3 L9 93613 .2 Q96 Qnd d.n.d.

WO 98/33919 PCT/US98/01826 92 EXAMPLE VII Generation of High Affinity LM609 Grafted Antibodies This example shows that single amino acid mutations in CDRs of an LM609 grafted that result in higher affinity binding to ov,3 can be combined to generate high affinity LM609 grafted antibodies.

Random combination of all of the beneficial mutations of LM609 grafted antibody would generate a combinatorial library containing >105 variants, requiring efficient screening methodologies. Therefore, to determine if clones displaying >10-fold enhanced affinities could be rapidly distinguished from one another, variants displaying 3 to 13-fold enhanced affinity were evaluated by capture lift utilizing lower concentrations of biotinylated a~03. Despite repeated attempts with a broad range of concentrations of aAv3, consistent differences in the capture lift signals were not observed. Because of this, smaller combinatorial libraries were constructed and subsequently screened by EL1SA.

Four distinct combinatorial libraries were constructed in order to evaluate the optimal number of combinations that could be accomplished utilizing two site hybridization mutagenesis (Figure Briefly, combinatorial libraries were constructed by synthesizing degenerate oligonucleotides encoding both the wild-type and beneficial heavy chain mutations (H2, Leu 60 -Pro; H3 Tyr 9 7 -His; H3, Ala 10 1 -Tyr; H3, Tyr' 02 -Ser, Thr, Asp, Glu, Met, Gly, Ala). Utilizing two site hybridization mutagenesis, as described above, the oligonucleotides were annealed at a 40:1 molar ratio to uridinylated template prepared from LM609 grafted antibody and three WO 98/33919 PCT/US98/01826 93 light chain mutations (Figure 9; L1, His 32 -Phe; L3, Gly 92 -Asn; L3, His 96 -Leu). As a result, a total of 256 variants were synthesized in four combinatorial library subsets.

Figure 9 shows construction of combinatorial libraries of beneficial mutations. Uridinylated template from LM609 grafted antibody and three optimal light chain variants (F32, N92, and L96) was prepared. Two site hybridization was performed with two degenerate oligonucleotides, which were designed to introduce beneficial mutations at four distinct heavy chain residues.

Following preparation of uridinylated templates of LM609 grafted antibody and three light chain variants, (Table 9; F32, N92, and L96), degenerate oligonucleotides encoding the wild type residue and the most beneficial heavy chain mutations (Table 9; P60, H97, Y101, S102, T102, D102, E102, M102, G102, and A102) were hybridized to the light chain templates, resulting in four combinatorial libraries, each containing 64 unique variants. Potentially, the combination of multiple mutations can have detrimental effects on affinity and, thus, can prevent the identification of beneficial combinations resulting from mutations at fewer sites.

For this reason, the amino acid expressed by the LM609 grafted antibody parent molecule was included at each position in the combinatorial library. By utilizing this approach, simultaneous combinatorial mutagenesis of three CDRs (L1 or L3 each in combination with H2 and H3) was accomplished. Based on sequence analysis, the two site hybridization mutagenesis was achieved with efficiency.

WO 98/33919 PCT/US98/01826 94 In order to screen the combinatorial libraries, soluble Fab was expressed and released from the periplasm of small-scale ml) bacterial cultures that had been infected with randomly selected clones. Although variable expression levels were observed, uniform quantities of the unpurified variants were captured on a microtiter plate through a peptide tag present on the carboxyl-terminus of the heavy chain. Briefly, combinatorial LM609 grafted antibody libraries were screened by an EL1SA that permits the determination of relative affinities of antibody variants produced in small-scale bacterial cultures (Watkins et al., Anal.

Biochem. 253:37-45 (1997)). An Immulon II microtiter plate (Dynatech Laboratories; Chantilly, VA) was coated with 10 ug/ml of the 7F11 monoclonal antibody, which recognizes a peptide tag on the carboxyl-terminus of the LM609 grafted antibody variant heavy chains (Field et al., Mol. Cell. Biol. 8:2159-2165 (1988)). Following capture of Fab from E. coli lysates, the plate was incubated with 0.5-1 ug/ml biotinylated c~v3 for 1 h at 250C. The plate was washed seven times, incubated with U/ml streptavidin-alkaline phosphatase (1000 U/ml; Boehringer Mannheim; Indianapolis, IN) for 15 min at 0 C, washed seven times, and developed as described previously (Watkins et al., supra, 1997). All dilutions and washes were in binding buffer.

As described previously (Watkins et al., supra, 1997), this ELISA screening method enabled a rapid and direct comparison of the relative affinities of the variants following incubation with biotinylated aP3 and streptavidin-alkaline phosphatase. To ensure that the full Fab diversity was sampled, one thousand randomly selected clones were screened from each combinatorial library. Variants that displayed an enhanced EL1SA WO 98/33919 PCT/US98/01826 signal were further characterized for binding to immobilized cVP 3 (Figure 8, closed triangles) and were sequenced to identify the mutations (Table SScreening of the four combinatorial libraries identified fourteen unique combinations of mutations that improved binding significantly over the individual mutations identified in the screening of the first library. While the best clone from the primary screen had a 12.5-fold increase in affinity, the fourteen unique combinations isolated from screening the combinatorial libraries displayed affinities ranging from 18 to 92-fold greater than the parent LM609 grafted antibody. The majority of these variants consisted of H2 and H3 mutations combined with the L1 or L3 mutations.

Beneficial combinations of heavy chain mutations with wild-type light chain were also identified, but did not result in improved affinity to the same extent as other combinatorial variants. The variants predominantly contained 2 to 4 mutations, with one clone, C29, containing five mutations. No direct correlation between the total number of mutations in each variant and the resulting affinity was observed. For example, while the binding of clone C37 was 92-fold enhanced over the parent molecule and was achieved through the combination of three mutations, clone C29 had -55-fold greater affinity achieved through the combination of five mutations.

Multiple variants displaying >50-fold enhanced affinity resulting from the combination of as few as two mutations were identified (2G4, 17, and V357D).

The combinatorial clones with improved affinity all displayed >10-fold slower dissociation rates, possibly reflecting a selection bias introduced by long incubation steps in the screening. In addition, all of WO 98/33919 PCT/US98/01826 96 the combinatorial variants isolated from the library based on the L96 light chain mutation also displayed 2 to 4-fold greater association rates. Previously, it has been demonstrated that the antibody repertoire shifts towards immunoglobulins displaying higher association rates during affinity maturation in vivo (Foote and Milstein, Nature 352:530-532 (1991)). The L96 subset of variants, therefore, may more closely mimick the in vivo affinity maturation process where B-lymphocyte proliferation is subject to a kinetic selection.

LM609 grafted antibody binds the a 3 complex specifically and does not recognize either the a, or the 3 3 chain separately. To further characterize the variants, clones were screened for reactivity with the related integrins, aI0,b3 and aP,5. All variants tested were unreactive with both a lbr 3 and YoP,, consistent with the improved binding not substantially altering the interaction of Fab and receptor.

C C C C C C SC *C C C C CCCC C C* CC *C* C. C. CCC C C C C C C C C *C CCC CC CC 97 Table 10: Identification of Optimal Combinatorial Mutations sequence t Li L3 L3 H2 H3 H3 H3 k_ (x10 4

K

0 (xl0-3) library* clone 32 92 96 60 97 101 102 Kd (nM) wild type H G H L y A Y 18.0 4.97 27.6 F32 17 F S 25.1 0.138 7 FP H S 20.4 0.236 1.2 56 F P S 26.6 0.135 C59 F P D 26.5 0.137 C176 F P T 22.5 0.192 0.9 V357D F D 27.9 0.140 N92 C119 N P S 21.5 0.316 L96 8F9 L P H S 47.5 0.280 0.6 C29 L P H Y S 67.5 0.343 2G4 L s 60.3 0.229 0.4 6H6 L H S 50.4 0.187 0.4 C37 L Y E 44.8 0.147 0.3 6D1 L P Y S 41.0 0.158 0.4 6G1 L P S 38.9 0.280 0.7 \\BRIS1\home$\IsabelH\Speci\34980 .doc WO 98/33919 PCT/US98/01826 98 As a first step toward determining if the increase in affinity of the variants resulted in greater biological activity, variants displaying a range of affinities were assayed for their ability to inhibit the binding of a natural ligand, fibrinogen, to immobilized av3 3 receptor. Briefly, LM609 grafted antibody variants were tested for inhibition of ligand binding as described previously except that the binding of biotinylated human fibrinogen (Calbiochem, La Jolla, CA) was detected with 0.5 ug/ml NeutrAvidin-alkaline phosphatase (Smith et al., J. Biol. Chem. 265:12267-12271 (1990)).

The results of these competition assays are shown in Figure 10. Figure 10A shows inhibition of fibrinogen binding to immobilized aUV 3 Immobilized 0~03 was incubated with 0.1 zg/ml biotinylated fibrinogen and various concentrations of LM609 grafted antibody (open circles), S102 (closed circles), F32 (open triangles), or C59 (closed triangles) for 3 h at 370C. Unbound ligand and Fab were removed by washing and bound fibrinogen was quantitated following incubation with NeutrAvidin alkaline phosphatase conjugate. Figure 10B shows correlation of affinity of variants with inhibition of fibrinogen binding. The concentration of variants required to inhibit the binding of fibrinogen to immobilized 0av 3 by 50% (IC 50 was plotted as a function of the affinity (Kd).

SAs shown in Figure 10A, higher affinity variants were more effective at blocking the ligand binding site of the receptor (compare LM609 grafted antibody, open circles, with any of the variants).

Subsequent analysis of ten variants displaying affinities (Kd) ranging from 0.3 to 27 nm demonstrated a good correlation (r 2 0.976) between affinity and ability to WO 98/33919 PCTUS98/01826 99 inhibit fibrinogen binding (Figure 10B). In addition, the variants were tested for inhibition of vitronectin binding to the receptor. Similar to fibrinogen, the variants were more effective at inhibiting the interaction than the parent molecule. Thus, consistent with the cross-reactivity studies with related integrin receptors, mutations which increased affinity did not appear to substantially alter the manner in which the antibody interacted with the receptor.

The ability of the variants to inhibit the adhesion of M21 human melanoma cells expressing the aA3 receptor to fibrinogen was examined. Inhibition of the adhesion of 4 x 104 M21 cells to fibrinogen by the LM609 grafted antibody variants was performed as described previously (Leavesley et al., J. Cell Biol. 117:1101-1107 (1992)). Similar to the ligand competition studies with purified fibrinogen and c v3 receptor, higher affinity variants were generally more effective at preventing cell adhesion than was LM609 grafted antibody (Figure 11) Figure 11 shows inhibition of M21 human melanoma cell adhesion to fibrinogen. Cells and various concentrations of LM609 grafted antibody Fab (closed triangles), S102 (open circles), G102 (closed circles), or C37 (open triangles) were added to 96 well cell culture plates which had been coated with 10 mg/ml fibrinogen. After incubating for 35 min at 37C, unbound cells were removed by washing and adherent cells were quantitated by crystal violet staining.

Although intact LM609 grafted antibody Ig inhibits cell adhesion, the phage expressed Fab did not affect cell adhesion at concentrations as high as 1 mg/ml (Figure 11, closed triangles). Clone C37, isolated from the combinatorial library and displaying -90-fold greater WO 98/33919 PCT/US98/01826 100 affinity than LM609 grafted antibody Fab, inhibited cell adhesion completely (Figure 11, open triangles). Variant G102 had a moderately higher affinity (2.2-fold enhanced) and also inhibited cell adhesion, though less effectively than C37 (Figure 11, closed circles). Surprisingly, clone S102 (Figure 11, open circles), which had a 4.6-fold higher affinity than LM609 grafted antibody, was ineffective at inhibiting cell adhesion, suggesting that clones G102 and S102 interact with the UAv3 receptor differently.

These results show that combining single amino acid mutations that result in LM609 grafted antibodies exhibiting higher binding affinity to UvA3 allows the identification of high affinity LM609 grafted antibody mutants having greater than 90-fold higher binding affinity than the parent LM609 grafted antibody.

Although the invention has been described with reference to the disclosed embodiments, those skilled in the art will readily appreciate that the specific experiments detailed are only illustrative of the invention. It should be understood that various modifications can be made without departing from the spirit of the invention. Accordingly, the invention is limited only by the following claims.

100a Throughout this specification and the claims, the words "comprise", "comprises" and "comprising" are used in a non-exclusive sense, except where the context requir.es otherwise.

It is to be clearly understood that any reference herein to a prior art publication does not constitute an admission that the document forms part of the common general knowledge in the art in Australia or in any other country.

e e e o• \\BRISl\home$\IsabelH\Speci\34980.doc

Claims (81)

1. A Vitaxin antibody exhibiting selective binding affinity to auv 3 comprising at least one Vitaxin heavy chain polypeptide comprising substantially the same variable region amino acid sequence as that shown in Figure 1A (SEQ ID NO:2) and at least one Vitaxin light chain polypeptide comprising substantially the same variable region amino acid sequence as that shown in Figure 1B (SEQ ID NO:4) or a functional fragment thereof, said antibody or functional fragment thereof having integrin avS3 binding activity, integrin-a~J 3 binding specificity or integrin aXvP 3 -inhibitory activity.

2. The Vitaxin antibody of claim 1, wherein said functional fragment is selected from the group consisting of Fv, Fab, F(ab) 2 and scFV. S3. A nucleic acid encoding a Vitaxin heavy chain polypeptide comprising substantially the same Vitaxin heavy chain variable region nucleotide sequences as that shown in Figure 1A (SEQ ID NO:1) or a fragment thereof.

4. The nucleic acid of claim 3, wherein said fragment further comprises a nucleic acid encoding S: substantially the same nucleotide sequence as the variable region of said Vitaxin heavy chain polypeptide (SEQ ID NO:1). The nucleic acid of claim 3, wherein said fragment further comprises a nucleic acid encoding substantially the same nucleotide sequence as a CDR of said Vitaxin heavy chain polypeptide. NU s S\\BRIS1\home$\IsabelH\Speci\34980.doc Sr 7 1 r 102

6. A nucleic acid encoding a Vitaxin light chain polypeptide comprising substantially the same Vitaxin light chain variable region nucleotide sequences as that shown in Figure 1B (SEQ ID NO:3) or a fragment thereof.

7. The nucleic acid of claim 6, wherein said fragment further comprises a nucleic acid encoding substantially the same nucleotide sequence as the variable region of said Vitaxin light chain polypeptide (SEQ ID NO:3).

8. The nucleic acid of claim 6, wherein said fragment further comprises a nucleic acid encoding substantially the same nucleotide sequence as a CDR of said Vitaxin light chain polypeptide. S.9. A nucleic acid encoding a Vitaxin heavy chain polypeptide comprising a nucleotide sequence encoding substantially the same Vitaxin heavy chain variable region amino acid sequence as that shown in Figure 1A (SEQ ID NO:2) or fragment thereof. S: 10. The nucleic acid of claim 9, wherein said fragment further comprises a nucleic acid encoding substantially the same heavy chain variable region amino acid sequence of said Vitaxin heavy chain amino acid sequence (SEQ ID NO:2).

11. The nucleic acid of claim 9, wherein said fragment further comprises a nucleic acid encoding substantially the same heavy chain CDR amino acid sequence of said Vitaxin heavy chain amino acid sequence.

12. A nucleic acid encoding a Vitaxin light chain polypeptide comprising a nucleotide sequence y \\BRISl\home$\IsabelH\Speci\34980.doc 103 encoding substantially the same Vitaxin light chain variable region amino acid sequence as that shown in Figure 1B (SEQ ID NO:4) or fragment thereof.

13. The nucleic acid of claim 12, wherein said fragment further comprises a nucleic acid encoding substantially the same light chain variable region amino acid sequence of said Vitaxin light chain amino acid sequence (SEQ ID NO:4).

14. The nucleic acid of claim 12, wherein said fragment further comprises a nucleic acid encoding substantially the same light chain CDR amino acid sequence of said Vitaxin light chain amino acid sequence.

15. A Vitaxin heavy chain polypeptide comprising substantially the same variable region amino acid sequence as that shown in Figure 1A (SEQ ID NO:2) or •functional fragment thereof.

16. The Vitaxin heavy chain polypeptide of claim 15, wherein said functional fragment comprises a variable chain polypeptide or a CDR polypeptide. ~17. A Vitaxin light chain polypeptide comprising substantially the same variable region amino acid sequence as that shown in Figure 1B (SEQ ID NO:4) or a functional fragment thereof.

18. The Vitaxin light chain polypeptide of claim 17, wherein said functional fragment comprises a variable chain polypeptide or a CDR polypeptide.

19. A LM609 grafted antibody exhibiting S 7' selective binding affinity to av3 comprising at least one SLM609 grafted heavy chain polypeptide comprising z \\BRISl\home$\IsabelH\Speci\34980.doc 104 substantially the same variable region amino acid sequence as that shown in Figure 1A (SEQ ID NO:2) and at least one LM609 grafted light chain polypeptide comprising substantially the same variable region amino acid sequence as that shown in Figure 7 (SEQ ID NO:32) or a functional fragment thereof said LM609 grafted antibody or functional fragment thereof having integrin aP3 3 binding activity, integrin Cv3 3 binding specificity or integrin av 3 -inhibitory activity. The LM609 grafted antibody of claim 19, wherein said functional fragment is selected from the group consisting of Fv, Fab, F(ab) 2 and scFV.

21. A nucleic acid encoding a LM609 grafted heavy chain polypeptide comprising substantially the same LM609 grafted heavy chain variable region nucleotide sequences as that shown in Figure 1A (SEQ ID NO:1) or a fragment thereof.

22. The nucleic acid of claim 21, wherein said S"fragment further comprises a nucleic acid encoding substantially the same nucleotide sequence as the variable region of said LM609 grafted heavy chain polypeptide (SEQ ID NO:1)

23. The nucleic acid of claim 21, wherein said fragment further comprises a nucleic acid encoding substantially the same nucleotide sequence as a CDR of said LM609 grafted heavy chain polypeptide.

24. A nucleic acid encoding a LM609 grafted light chain polypeptide comprising substantially the same LM609 grafted light chain variable region nucleotide \\BRISl\home$\IsabelH\Speci\34980.doc 105 sequences as that shown in Figure 7 (SEQ ID NO:31) or a fragment thereof. The nucleic acid of claim 24, wherein said fragment further comprises a nucleic acid encoding substantially the same nucleotide sequence as the variable region of said LM609 grafted light chain polypeptide (SEQ ID NO:31)

26. The nucleic acid of claim 24, wherein said fragment further comprises a nucleic acid encoding substantially the same nucleotide sequence as a CDR of said LM609 grafted light chain polypeptide.

27. A nucleic acid encoding a LM609 grafted •antibody heavy chain polypeptide comprising a nucleotide sequence encoding substantially the same LM609 grafted heavy chain variable region amino acid sequence as that shown in Figure 1A (SEQ ID NO:2) or fragment thereof.

28. The nucleic acid of claim 27, wherein said fragment further comprises a nucleic acid encoding substantially the same heavy chain variable region amino acid sequence of said LM609 grafted heavy chain amino acid sequence (SEQ ID NO:2).

29. The nucleic acid of claim 27, wherein said 9 fragment further comprises a nucleic acid encoding substantially the same heavy chain CDR amino acid sequence of said LM609 grafted heavy chain amino acid sequence. A nucleic acid encoding a LM609 grafted antibody light chain polypeptide comprising a nucleotide sequence encoding substantially the same LM609 grafted \\BRISl\home$\IsabelH\Speci\34980.doc 106 light chain variable region amino acid sequence as that shown in Figure 7 (SEQ ID NO:32) or fragment thereof.

31. The nucleic acid of claim 30, wherein said fragment further comprises a nucleic acid encoding substantially the same light chain variable region amino acid sequence of said Vitaxin light chain amino acid sequence (SEQ ID NO:32).

32. The nucleic acid of claim 30, wherein said fragment further comprises a nucleic acid encoding substantially the same light chain CDR amino acid sequence of said Vitaxin light chain amino acid sequence.

33. A LM609 grafted heavy chain polypeptide comprising substantially the same variable region amino acid sequence as that shown in Figure 1A (SEQ ID NO:2) or functional fragment thereof. S: 34. The LM609 grafted heavy chain polypeptide of claim 33, wherein said functional fragment comprises a variable chain polypeptide or a CDR polypeptide. A LM609 grafted light chain polypeptide "comprising substantially the same variable region amino acid sequence as that shown in Figure 7 (SEQ ID NO:32) or a functional fragment thereof.

36. The LM609 grafted light chain polypeptide of claim 35, wherein said functional fragment comprises a variable chain polypeptide or a CDR polypeptide.

37. A nucleic acid encoding a heavy chain cS polypeptide for monoclonal antibody LM609 comprising substantially the same heavy chain variable region \\BRISl\home$\IsabelH\Speci\34980.doc 107 nucleotide sequence as that shown in Figure 2A (SEQ ID or a fragment thereof.

38. The nucleic acid of claim 37, wherein said fragment further comprises a nucleic acid encoding substantially the same nucleotide sequence as the variable region of said heavy chain polypeptide (SEQ ID

39. The nucleic acid of claim 37, wherein said fragment further comprises a nucleic acid encoding substantially the same nucleotide sequence as a CDR of said heavy chain polypeptide. A nucleic acid encoding a light chain polypeptide for monoclonal antibody LM609 comprising substantially the same light chain variable region nucleotide sequence as that shown in Figure 2B (SEQ ID NO:7) or a fragment thereof.

41. The nucleic acid of claim 40, wherein said fragment further comprises a nucleic acid encoding substantially the same nucleotide sequence as the variable region of said light chain polypeptide (SEQ ID NO:7). a

42. The nucleic acid of claim 40, wherein said f Sf fragment further comprises a nucleic acid encoding substantially the same nucleotide sequence as a CDR of said light chain polypeptide.

43. A nucleic acid encoding a heavy chain polypeptide for monoclonal antibody LM609 comprising a nucleotide sequence encoding substantially the same heavy f chain variable domain amino acid sequence of monoclonal \\BRISl\home$\IsabelH\Speci\34980.doc 108 antibody LM609 as that shown in Figure 2A (SEQ ID NO:6) or fragment thereof.

44. The nucleic acid of claim 43, wherein said fragment further comprises a nucleic acid encoding substantially the same heavy chain variable region amino acid sequence of said monoclonal antibody LM609 (SEQ ID NO:6). The nucleic acid of claim 43, wherein said fragment further comprises a nucleic acid encoding substantially the same heavy chain CDR amino acid sequence as said monoclonal antibody LM609.

46. A nucleic acid encoding a heavy chain polypeptide for monoclonal antibody LM609 comprising a nucleotide sequence encoding substantially the same light chain amino acid sequence of monoclonal antibody LM609 as that shown in Figure 2B (SEQ ID NO:8) or fragment **thereof.

47. The nucleic acid of claim 46, wherein said fragment further comprises a nucleic acid encoding substantially the same light chain variable region amino acid sequence of said monoclonal antibody LM609 (SEQ ID NO:8).

48. The nucleic acid of claim 46, wherein said fragment further comprises a nucleic acid encoding substantially the same light chain CDR amino acid sequence as said monoclonal antibody LM609.

49. Use of Vitaxin or a functional fragment thereof for the preparation of a pharmaceutical composition for inhibiting a function of a,3. \\BRISl\home$\IsabelH\Speci\34980.doc 109 The use of claim 49, wherein said functional fragment is selected from the group consisting of Fv, Fab, F(ab) 2 and scFV.

51. The use of claim 49, wherein said function of vI3v is binding of av3 to a ligand.

52. The use of claim 49, wherein said function of av 3 3 is integrin mediated signal transduction.

53. Use of Vitaxin or a functional fragment thereof for the preparation of a pharmaceutical composition for the treatment of an a,03-mediated disease.

54. The use of claim 53, wherein said functional fragment is selected from the group consisting of Fv, Fab, F(ab) 2 and scFV. The use of claim 53, wherein said vPI3-mediated disease is angiogenesis or restenosis.

56. An enhanced LM609 grafted antibody exhibiting selective binding affinity to avP 3 or a functional fragment thereof, comprising one or more CDRs having at least one amino acid substitution in one or more CDRs of a LM609 grafted heavy chain variable region polypeptide referenced as SEQ ID NO:6 or a LM609 grafted light chain variable region polypeptide referenced as SEQ ID NO:8, said antibody or functional fragment thereof having integrin 01j 3 binding activity, integrin av3 3 binding specificity or integrin cv3 3 -inhibitory activity, wherein the C, 3 3 binding affinity of said enhanced LM609 grafted antibody is maintained. C F \\BRISl\home$\IsabelH\Speci\34980.doc o i 110

57. The enhanced LM609 grafted antibody of claim 56, wherein said av3 binding affinity of said LM609 grafted antibody is enhanced.

58. The enhanced LM609 grafted antibody of claim 56, wherein said functional fragment is selected from the group consisting of Fv, Fab, F(ab) 2 and scFV.

59. The enhanced LM609 grafted antibody of claim 56, wherein said CDR having at least one amino acid substitution is selected from the group consisting of VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3. The enhanced LM609 grafted antibody of claim 59, wherein said VH CDR1 is selected from the group consisting of the CDRs referenced as SEQ ID NO:48, SEQ ID NO:50 and SEQ ID NO:52.

61. The enhanced LM609 grafted antibody of claim 59, wherein said VH CDR2 is selected from the group consisting of the CDRs referenced as SEQ ID NO:54, SEQ ID NO:56 and SEQ ID NO:58.

62. The enhanced LM609 grafted antibody of claim 59, wherein said VH CDR3 is selected from the group consisting of the CDRs referenced as SEQ ID NO:60, SEQ ID NO:62, SEQ ID NO:64, SEQ ID NO:66, SEQ ID NO:68, SEQ ID SEQ ID NO:72, SEQ ID NO:74, SEQ ID NO:76, SEQ ID NO:78, SEQ ID NO:72, SEQ ID NO:74, SEQ ID NO:76; SEQ ID N0:78, SEQ ID NO:80, SEQ ID NO:94, SEQ ID NO:96; SEQ ID NO:98 and SEQ ID NO:100.

63. The enhanced LM609 grafted antibody of claim 59, wherein said VL CDR1 is the CDR referenced as SEQ ID NO:82. \\BRISl\home$\IsabelH\Speci\34980.doc 111

64. The enhanced LM609 grafted antibody of claim 59, wherein said VL CDR2 is the CDR referenced as SEQ ID NO:84. The enhanced LM609 grafted antibody of claim 59, wherein said VL CDR3 is selected from the group consisting of the CDRs referenced as SEQ ID NO:86, SEQ ID NO:88, SEQ ID NO:90 and SEQ ID NO:92.

66. The enhanced LM609 grafted antibody of claim 56, wherein said enhanced LM609 grafted antibody comprises at least one amino acid substitution in two or more CDRs of a LM609 grafted heavy chain variable region polypeptide or a LM609 grafted light chain variable region polypeptide.

67. The enhanced LM609 grafted antibody of claims 66, wherein said functional fragment is selected from the group consisting of Fv, Fab, F(ab) 2 and scFV.

68. The enhanced LM609 grafted antibody of claim 66, wherein said CDR having at least one amino acid substitution is selected from the group consisting of VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3.

69. The enhanced LM609 grafted antibody of claim 68, wherein said enhanced LM609 grafted antibody comprises the combination of CDRs selected from the group consisting of: the VL CDR1 referenced as SEQ ID NO:82 and the VH CDR3 referenced as SEQ ID NO:68; the VL CDRl referenced as SEQ ID NO:82, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID NO:68; \\BRISl\home$\IsabelH\Speci\34980.doc 112 the V, CDR1 referenced as SEQ ID NO:82, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID NO:72; the VL CDR1 referenced as SEQ ID NO:82, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID the VL CDR1 referenced as SEQ ID NO:82 and the VH CDR3 referenced as SEQ ID NO:72; the VL CDR3 referenced as SEQ ID NO:86, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID NO:68; the VL CDR3 referenced as SEQ ID NO:90 and VH CDR3 referenced as SEQ ID NO:68; and the VL CDR3 referenced as SEQ ID NO:90, the VH CDR2 referenced as SEQ ID NO:56 and VH CDR3 referenced as SEQ ID NO:68.

70. The enhanced LM609 grafted antibody of 9 claim 66, wherein at least one of said CDRs has two or more amino acid substitutions.

71. The enhanced LM609 grafted antibody of claims 70, wherein said functional fragment is selected from the group consisting of Fv, Fab, F(ab) 2 and scFV.

72. The enhanced LM609 grafted antibody of claim 70, wherein said CDR having at least one amino acid substitution is selected from the group consisting of VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3.

73. The enhanced LM609 grafted antibody of claim 72, wherein said enhanced LM609 grafted antibody comprises the combination of CDRs selected from the group consisting of: \\BRISl\home$\IsabelH\speci\34980.doc 113 the VL CDR1 referenced as CDR2 referenced as SEQ ID NO:56 and as SEQ ID NO:94; the VL CDR3 referenced as CDR2 referenced as SEQ ID NO:56 and as SEQ ID NO:94; the VL CDR3 referenced as CDR2 referenced as SEQ ID NO:56 and as SEQ ID NO:96; the VL CDR3 referenced as VH CDR3 referenced as SEQ ID NO:94; the VL CDR3 referenced as VH CDR3 referenced as SEQ ID NO:98; the VL CDR3 referenced as CDR2 referenced as SEQ ID NO:56 and as SEQ ID NO:100. SEQ ID NO:82, the VH the VH CDR3 referenced SEQ ID NO:90, the VH the VH CDR3 referenced SEQ ID NO:90, the VH the VH CDR3 referenced SEQ ID NO:90 and the SEQ ID NO:90 and the and SEQ ID NO:90, the VH the VH CDR3 referenced 0 S S S 6* S* OS *r 0 0 S 0. S *0*S S a.r 0 S 60* 0

74. A high affinity LM609 grafted antibody exhibiting selective binding affinity to a(v3, or a functional fragment thereof, comprising one or more CDRs having at least one amino acid substitution in one or more CDRs of a LM609 grafted heavy chain variable region polypeptide referenced as SEQ ID NO:6 or a LM609 grafted light chain variable region polypeptide referenced as SEQ ID NO:8, said antibody or functional fragment thereof having integrin avP 3 binding activity, integrin avP3 binding specificity or integrin avP3-inhibitory activity, wherein the 0a33 binding affinity of said high affinity LM609 grafted antibody is enhanced. The high affinity LM609 grafted antibody of claim 74, wherein said functional fragment is selected from the group consisting of Fv, Fab, F(ab) 2 and scFV. \\BRISl\home$\IsabelH\Speci\34980.doc 114

76. The high affinity LM609 grafted antibody of claim 74, wherein said CDR having at least one amino acid substitution is selected from the group consisting of VH CDR1, VH CDR2, VH CDR3, VL CDR1, VL CDR2 and VL CDR3.

77. The high affinity LM609 grafted antibody of claim 76, wherein said high affinity LM609 grafted antibody comprises the combination of CDRs selected from the group consisting of: *0 *0@O S* I. S. O OO fo 0 S *C C C ec C OCS 0 the VL CDR1 referenced as SEQ VH CDR3 referenced as SEQ ID NO:68; the VL CDR1 referenced as SEQ CDR2 referenced as SEQ ID NO:56 and the as SEQ ID NO:68; the VL CDR1 referenced as SEQ CDR2 referenced as SEQ ID NO:56 and the as SEQ ID NO:72; the VL CDR1 referenced as SEQ CDR2 referenced as SEQ ID NO:56 and the as SEQ ID the VL CDR1 referenced as SEQ VH CDR3 referenced as SEQ ID NO:72; the VL CDR3 referenced as SEQ CDR2 referenced as SEQ ID NO:56 and the as SEQ ID NO:68; the VL CDR3 referenced as SEQ CDR2 referenced as SEQ ID NO:56 and the as SEQ ID NO:94; the VL CDR3 referenced as SEQ CDR3 referenced as SEQ ID NO:68; the VL CDR3 referenced as SEQ ID NO:82 and the ID NO:82, the VH VH CDR3 referenced ID NO:82, the VH VH CDR3 referenced ID NO:82, the VH VH CDR3 referenced ID NO:82 and the ID NO:86, the VH VH CDR3 referenced ID NO:90, the VH VH CDR3 referenced ID NO:90 and VH ID NO:90, the VH CDR2 referenced as SEQ ID NO:56 and VH CDR3 referenced as SEQ ID NO:68; \\BRISl\home$\IsabelH\Speci\34980.doc 115 the VL CDR1 referenced as SEQ ID NO:82, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID NO:94; the VL CDR3 referenced as SEQ ID NO:90, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID NO:96; the VL CDR3 referenced as SEQ ID NO:90 and the VH CDR3 referenced as SEQ ID NO:94; the VL CDR3 referenced as SEQ ID NO:90 and the VH CDR3 referenced as SEQ ID NO:98; and the VL CDR3 referenced as SEQ ID NO:90, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID NO:100.

78. A nucleic acid encoding an enhanced LM609 grafted antibody, or a functional fragment thereof, exhibiting selective binding affinity to va3 3 comprising at least one amino acid substitution in one or more CDRs of a LM609 grafted heavy chain variable region polypeptide or a LM609 grafted light chain variable region polypeptide, antibody or functional fragment thereof having integrin (3P3 binding activity, integrin va3 3 binding specificity or integrin avP3-inhibitory activity, wherein the aP3 binding affinity of said enhanced LM609 grafted antibody is maintained or enhanced. eo

79. A nucleic acid encoding a high affinity LM609 grafted antibody, or a functional fragment thereof, exhibiting selective binding affinity to a~3 3 comprising at least one amino acid substitution in one or more CDRs of a LM609 grafted heavy chain variable region polypeptide or a LM609 grafted light chain variable region polypeptide, wherein the a vj binding affinity of ST said high affinity LM609 grafted antibody is enhanced. \\BRISl\home$\IsabelH\Speci\34980.doc 116 An antibody, or functional fragment thereof, comprising a VH CDR1 selected from the group consisting of the CDRs referenced as SEQ ID NO:48, SEQ ID and SEQ ID NO:52.

81. The antibody of claim 80, wherein said functional fragment is selected from the group consisting *of Fv, Fab, F(ab) 2 and scFV.

82. An antibody, or functional fragment thereof, comprising a VH CDR2 selected from the group consisting of the CDRs referenced as SEQ ID NO:54, SEQ ID NO:56 and SEQ ID NO:58.

83. The antibody of claim 82, wherein said functional fragment is selected from the group consisting of Fv, Fab, F(ab) 2 and scFV. 0

84. An antibody, or functional fragment thereof, comprising a VH CDR3 selected from the group consisting of the CDRs referenced as SEQ ID NO:60, SEQ ID NO:62, SEQ ID NO:64, SEQ ID NO:66, SEQ ID NO:68, SEQ ID SEQ ID NO:72, SEQ ID NO:74, SEQ ID NO:76, SEQ ID NO:78, SEQ ID NO:80, SEQ ID NO:94, SEQ ID NO:96; SEQ ID NO:98 and SEQ ID NO:100. *S S: 85. The antibody of claim 84, wherein said functional fragment is selected from the group consisting of Fv, Fab, F(ab) 2 and scFV.

86. An antibody, or functional fragment thereof, comprising a VL CDR1 referenced as SEQ ID NO:82. T87. The antibody of claim 86, wherein said functional fragment is selected from the group consisting \\BRISl\home$\IsabelH\Speci\34980.doc 117 of Fv, Fab, F(ab) 2 and scFV.

88. An antibody, or functional fragment thereof, comprising a VL CDR2 referenced as SEQ ID NO:84.

89. The antibody of claim 88, wherein said functional fragment is selected from the group consisting of Fv, Fab, F(ab) 2 and scFV. An antibody, or functional fragment thereof, comprising a VL CDR3 selected from the group consisting of the CDRs referenced as SEQ ID NO:86, SEQ ID NO:88, SEQ ID NO:90 and SEQ ID NO:92.

91. The antibody of claim 90, wherein said functional fragment is selected from the group consisting of Fv, Fab, F(ab) 2 and scFV.

92. An antibody, or functional fragment thereof, comprising the combination of CDRs selected from the group consisting of: SC3 the VL CDR1 referenced as SEQ ID NO:82 and the VH CDR3 referenced as SEQ ID NO:68; the VL CDR1 referenced as SEQ ID NO:82, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID NO:68; the VL CDR1 referenced as SEQ ID NO:82, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID NO:72; the VL CDR1 referenced as SEQ ID NO:.82, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID the VL, CDR1 referenced as SEQ ID NO:82 and the Vui CDR3 referenced as SEQ ID NO:72; i/ the VL CDR3 referenced as SEQ ID NO:86, the VH U z \\BRISl\home$\IsabelH\Speci\34980.doc 118 CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID NO:68; the VL CDR3 referenced as SEQ ID NO:90 and VH CDR3 referenced as SEQ ID NO:68; and the VL CDR3 referenced as SEQ ID NO:90, the VH CDR2 referenced as SEQ ID NO:56 and VH CDR3 referenced as SEQ ID NO:68.

93. The antibody of claims 92, wherein said functional fragment is selected from the group consisting of Fv, Fab, F(ab) 2 and scFV.

94. An antibody, or functional fragment thereof, comprising the combination of CDRs selected from the group consisting of: C C CC C C. the VL CDR1 referenced as SEQ ID CDR2 referenced as SEQ ID NO:56 and the VH as SEQ ID NO:94; the VL CDR3 referenced as SEQ ID CDR2 referenced as SEQ ID NO:56 and the VH as SEQ ID NO:94; the VL CDR3 referenced as SEQ ID CDR2 referenced as SEQ ID NO:56 and the VH as SEQ ID NO:96; the VL CDR3 referenced as SEQ ID VH CDR3 referenced as SEQ ID NO:94; the VL CDR3 referenced as SEQ ID VH CDR3.referenced as SEQ ID NO:98; and the VL CDR3 referenced as SEQ ID CDR2 referenced as SEQ ID NO:56 and the VH as SEQ ID NO:100. NO:82, the VH CDR3 referenced NO:90, the VH CDR3 referenced NO:90, the VH CDR3 referenced NO:90 and the NO:90 and the NO:90, the VH CDR3 referenced The antibody of claims 94, wherein said functional fragment is selected from the group consisting of Fv, Fab, F(ab) 2 and scFV. \\BRISl\home$\IsabelH\Speci\34980.doc 119

96. An antibody, or functional fragment thereof, comprising the combination of CDRs selected from the group consisting of: the VL CDR1 referenced as SEQ ID NO:82 and the VH CDR3 referenced as SEQ ID NO:68; the VL CDR1 referenced as SEQ ID NO:82, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID NO:68; the VL CDR1 referenced as SEQ ID NO:82, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID NO:72; the VL CDR1 referenced as SEQ ID NO:82, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID the VL CDR1 referenced as SEQ ID NO:82 and the VH CDR3 referenced as SEQ ID NO:72; the VL CDR3 referenced as SEQ ID NO:86, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID NO:68; the VL CDR3 referenced as SEQ ID NO:90, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced :as SEQ ID NO:94; the VL CDR3 referenced as SEQ ID NO:90 and VH CDR3 referenced as SEQ ID NO:68; the VL CDR3 referenced as SEQ ID NO:90, the VH SCDR2 referenced as SEQ ID NO:56 and VH CDR3 referenced as SEQ ID NO:68; the VL CDR1 referenced as SEQ ID NO:82, the VH CDR2. referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID NO:94; the VL CDR3 referenced as SEQ ID NO:90, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID NO:96; STF& the VL CDR3 referenced as SEQ ID NO:90 and the \\BRISl\home$\IsabelH\Speci\34980.doc 120 VH CDR3 referenced as SEQ ID NO:94; the VL CDR3 referenced as SEQ ID NO:90 and the VH CDR3 referenced as SEQ ID NO:98; and the VL CDR3 referenced as SEQ ID NO:90, the VH CDR2 referenced as SEQ ID NO:56 and the VH CDR3 referenced as SEQ ID NO:100.

97. The antibody of claim 96, wherein said functional fragment is selected from the group consisting of Fv, Fab, F(ab) 2 and scFV.

98. A method of inhibiting a function of caj3 comprising contacting avy3 with the enhanced LM609 grafted antibody of claim 56, or a functional fragment thereof, under conditions which allow binding of said enhanced LM609 grafted antibody to af3,.

99. The method of claim 98, wherein said functional fragment is selected from the group consisting of Fv, Fab, F(ab) 2 and scFV.

100. The method of claim 98, wherein said function of caf, is binding of avf, to a ligand. 1.01. The method of claim 98, wherein said function of a 3 is integrin mediated signal transduction.

102. A method of treating an ar,-mediated disease comprising administering an effective amount of the enhanced LM609 grafted antibody of claim 56, or a functional fragment thereof, under conditions which allow binding to a.

103. The method of claim 102, wherein said Sfunctional fragment is selected from the group consisting Z H:\IsabelH\Speci\34980.doc 121 of Fv, Fab, F(ab) 2 and scFV.

104. The method of claim 102, wherein said aP 3 mediated disease is angiogenesis or restenosis. Dated this 21st day of June 2001 IXSYS, INCORPORATED By their Patent Attorneys GRIFFITH HACK 4 .44444 .4 4 44*4 44.44 4 44.4 p 4*44 4. *4 4 4 4 .44. a 4.. 4 4* 4 4* 4 a H: \IsabelH\Speci\34980 .doc